Policy Analysis in the Netherlands 9781447313342

Citation preview

INTERNATIONAL LIBRARY OF POLICY ANALYSIS SERIES EDITORS: IRIS GEVA-MAY & MICHAEL HOWLETT

POLICY ANALYSIS IN

The Netherlands

Edited by Frans van Nispen and Peter Scholten

POLICY ANALYSIS IN THE NETHERLANDS

International Library of Policy Analysis Series editors: Iris Geva-May and Michael Howlett, Simon Fraser University, Canada This major new series brings together for the first time a detailed examination of the theory and practice of policy analysis systems at different levels of government and by non-governmental actors in a specific country. It therefore provides a key addition to research and teaching in comparative policy analysis and policy studies more generally.  Each volume includes a history of the country’s policy analysis which offers a broad comparative overview with other countries as well as the country in question. In doing so, the books in the series provide the data and empirical case studies essential for instruction and for further research in the area. They also include expert analysis of different approaches to policy analysis and an assessment of their evolution and operation. Early volumes in the series will cover the following countries: Brazil • Germany • Netherlands • Japan • Taiwan • Israel • Australia • Czech Republic • Belgium • France and will build into an essential library of key reference works. The series will be of interest to academics and students in public policy, public administration and management, comparative politics and government, public organisations and individual policy areas. It will also interest people working in the countries in question and internationally. In association with the ICPA-Forum and Journal of Comparative Policy Analysis. See more at http://goo.gl/raJUX

POLICY ANALYSIS IN THE NETHERLANDS Edited by Frans van Nispen and Peter Scholten

International Library of Policy Analysis, Vol 3

First published in Great Britain in 2015 by Policy Press North America office: University of Bristol Policy Press 1-9 Old Park Hill c/o The University of Chicago Press Bristol BS2 8BB 1427 East 60th Street UK Chicago, IL 60637, USA t: +44 (0)117 954 5940 t: +1 773 702 7700 [email protected] f: +1 773-702-9756 www.policypress.co.uk [email protected] www.press.uchicago.edu © Policy Press 2015 British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication Data A catalog record for this book has been requested. ISBN 978 1 44731 333 5 hardcover The right of Frans van Nispen and Peter Scholten to be identified as editors of this work has been asserted by them in accordance with the 1988 Copyright, Designs and Patents Act. All rights reserved: no part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise without the prior permission of Policy Press. The statements and opinions contained within this publication are solely those of the editors and contributors and not of The University of Bristol or Policy Press. The University of Bristol and Policy Press disclaim responsibility for any injury to persons or property resulting from any material published in this publication. Policy Press works to counter discrimination on grounds of gender, race, disability, age and sexuality. Cover design by Qube Design Associates, Bristol Front cover: image kindly supplied by www.istock.com Printed and bound in Great Britain by CPI Group (UK) Ltd, Croydon, CR0 4YY Policy Press uses environmentally responsible print partners.

Contents List of figures and tables List of acronyms Notes on contributors Editors’ introduction to the series Preface and acknowledgements One

Policy analysis in the Netherlands: an introduction Frans van Nispen and Peter Scholten 1.1 Introduction 1.2 Policy analysis in Dutch academia 1.3 Policy analysis in the Dutch polity 1.4 A conceptual map of policy analysis in the Netherlands and beyond 1.5 Outline of this book

Part One: Policy styles and modes of policy analysis Two Policy analysis in practice: reinterpreting the quest for evidence-based policy Mark van Twist, Rien Rouw and Martijn van der Steen 2.1 Introduction 2.2 Theoretical debate on evidence-based policy 2.3 An empirical perspective: the practical application of knowledge in policy? 2.4 Conclusion and discussion: towards deliberative evidence-based policy Three Four

Policy analysis in networks: the battle of analysis and the potentials of joint fact-finding Arwin van Buuren and Joop Koppenjan 3.1 Introduction 3.2 Knowledge for governance in complex networks: joint fact-fighting or joint fact-finding? 3.3 The governance of knowledge production and knowledge use in governance networks 3.4 Case analysis: enhancing retention capacity in the Dutch Gouwe Wiericke polder 3.5 Analysis: distance and co-production, shifting accents 3.6 Explaining why knowledge and the solutions based upon it did not become authoritative 3.7 Conclusions: organising co-production and distance in the triangle Patterns of science–policy interaction Robert Hoppe 4.1 Introduction: science-informed or expert policy advice in the Netherlands

ix x xiv xx xxii 1 1 2 4 5 9

17 17 20 24 26 33 33 34 36 38 41 44 45 49 49

v

Policy analysis in the Netherlands Five

4.2 A minimalist framework of expert policy advice as boundary work 4.3 Boundary arrangements, organisations and policy politics in expert advice 4.4 Political cultures and policy styles in the Dutch advisory infrastructure 4.5 Changes in the problem and the polity, changes in science-informed policy advice

50

Interpretive policy analysis in the Netherlands Severine van Bommel, Merlijn van Hulst and Dvora Yanow 5.1 Introduction 5.2 An interpretive approach to policy research 5.3 Developing interpretive policy analysis in the Netherlands: a genealogy 5.4 Main theoretical-analytic focuses of Dutch interpretive policy research 5.5 Understanding interpretive policy analysis in the Netherlands

69

Part Two: Policy analysis in government Six Policy analysis and evaluation in national government Frans van Nispen 6.1 Introduction 6.2 The Committee for the Development of Policy Analysis (1970–81) 6.3 The shotgun marriage between policy analysis and public budgeting 6.4 The Reconsideration of Public Expenditures 6.5 Policy reviews (from 2004 onwards) 6.6 Conclusion Seven Eight

vi

Policy analysis at the local government level Frank Hilterman and Henk Klaassen 7.1 Introduction 7.2 The Dutch political system 7.3 Rationalisation of public policy at the local level: some institutional aspects 7.4 Towards New Public Management 7.5 Policy and management instruments 7.6 The Dutch approach to performance-based budgeting 7.7 Bottom-up policy analysis 7.8 Evaluation and critical remarks Policy analysis and performance audit at the ‘highest level’: looking for evidence and responsiveness Peter van der Knaap 8.1 Introduction 8.2 Policy analysis and policy evaluation in the Netherlands at the highest level: setting the stage

52 56 62

69 69 72 74 78

89 89 90 91 95 99 101 107 107 107 108 111 112 116 117 118 121 121 121

Contents

8.3 Policy analysis at the Council of State: ex ante advice on new policies 8.4 Policy analysis and performance audits at the Court of Audit: moving towards responsive ‘reality checks’ 8.5 Policy analysis and the National Ombudsman 8.6 New directions: evidence and responsiveness

Part Three: Advisory boards, consultancy firms, research institutes and think tanks Nine Advisory boards and planning bureaus Peter Scholten and Frans van Nispen 9.1 Introduction 9.2 Origin and development of the Dutch system of advisory bodies 9.3 Key advisory bodies in the Netherlands 9.4 Conclusion: Dutch advisory bodies in a broader political administrative setting Ten

The Dutch policy research industry Peter van Hoesel 10.1 Introduction 10.2 Policy research providers 10.3 The demand for external policy research 10.4 Invitation to tender for research projects 10.5 Programming policy research 10.6 Supervising research projects 10.7 Dissemination and utilisation of research findings 10.8 Conclusion

Part Four: Policy analysis in politics and by interest groups in society Eleven Policy analysis in the Dutch Parliament Karin Zaal 11.1 Introduction 11.2 Position and role of the Dutch House of Representatives 11.3 The House of Representatives and its access to information 11.4 Policy analysis in the House of Representatives: a practical guide 11.5 Conclusions Twelve

Policy analysis and political party think tanks Arco Timmermans, Edwin van Rooyen and Gerrit Voerman 12.1 Introduction 12.2 The emergence of party think tanks 12.3 The party think-tank agenda 12.4 Self-reported experiences of party think tanks 11.5 Conclusion

124 127 132 134

139 139 140 142 151 155 155 156 159 160 162 163 163 165

171 171 171 174 179 184 187 187 188 190 196 197

vii

Policy analysis in the Netherlands Thirteen Policy analysis by negotiation: trade unions, employers’ organisations and the Social and Economic Council Menno Fenger and Pierre Koning 13.1 Introduction 13.2 The institutional landscape of the Dutch polder model 13.3 Negotiated knowledge as policy analysis 13.4 Case One – reforming Disability Insurance 13.5 Case Two – reforming pensions 13.6 Conclusions Part Five: Policy analysis in the academic world Fourteen Policy analysis as instruction Wil Thissen 14.1 Introduction 14.2 Public policy and management education in the Netherlands 14.3 Policy analysis in education 14.4 The Delft University of Technology programme 14.5 Issues and experiences in education 14.6 Closing remarks and reflections Fifteen

Contested knowledge in theory-driven policy analysis: setting the Dutch stage Victor Bekkers 15.1 Introduction 15.2 Four perspectives on (the role of knowledge in) policymaking and analysis 15.3 Perspectives on policy analysis and the role of knowledge in the Netherlands 15.4 Three emerging trends 15.5 Three conflicting knowledge reservoirs 15.6 Towards a conclusion

203 203 204 206 208 210 211

217 217 218 220 221 226 227 231 231 232 233 237 238 242

Part Six: Conclusion Sixteen Institutionalisation and performance of policy analysis and evaluation in the Netherlands Peter Scholten and Frans van Nispen 16.1 Introduction 16.2 Institutionalisation 16.3 Policy performance 16.4 Academic performance 16.5 Looking back and forth

247 248 249 251 252

Index

255

viii

247

Policy analysis in the Netherlands

List of figures and tables List of figures 1.1 3.1 4.1 4.2 4.3 6.1 6.2 6.3 6.4 6.5 6.6 6.7 7.1 7.2 8.1 9.1 10.1 10.2 12.1

A framework for policy analysis 9 Map of Gouwe Wiericke and De Venen 40 Multi-level conceptual framework for understanding science–policy 51 interactions Problem types and types of policy politics 52 Problem types and some exemplary advisory bodies in the Netherlands 56 Budget template 92 The coverage of outputs by performance indicators 93 The revision of the WWW-questions 94 A typology of policy analysis 95 The adoption of potential savings per year 97 The utilisation of the BHO in the election manifestos of the political 98 parties (2010) The utilisation of the BHO in the Coalition Agreement of the Rutte I 99 Cabinet (2010) BAG Model I 110 BAG Model II 111 How policy advice is generated and used: policy analysis and 123 performance auditing in the central government policy process The budgetary consequences of the election manifestos of the political 150 parties The size of the budget per client 160 The size of the budget per type of policy research 161 Allocation of attention on think-tank agendas 195

List of tables 1.1 1.2 3.1 3.2 5.1 6.1 7.1 7.2 8.1 8.1 11.1 12.1 14.1

ix

Policy analysis versus policy research or studies A classification of public policy Joint fact-finding in the case of Gouwe Wiericke Co-production and distance in three rounds Five Dutch interpretive policy-analysis approaches Modes of policy analysis and evaluation New Public Management in Dutch municipalities BBI: phases, premises and goals The three institutions compared A survey of four main advisory boards of the Dutch government Attention to ideology and party principles (average percentages per government term) Conditions for negotiated knowledge to work – the two case studies compared Overview of policy-oriented educational programmes in the Netherlands

6 8 44 45 78 100 113 113 134 142 195 212 219

Policy analysis in the Netherlands

List of acronyms Acronym ACM

In full Autoriteit Consument & Markt

ARK

Algemene Rekenkamer

ARP

Anti-Revolutionaire Partij

AOV AZ B&M BAG BBI BBV BHO BOR CBA CBS CDA CEP CHU COBA CPB CRM CSR CSV CU CWOK D66

x

English translation Authority for Consumers and Markets Netherlands Court of Audit (NCA)

Anti-Revolutionary Party [one of the predecessors of the CDA] Arbeidsongeschiktheidsverzekering Disability Insurance Ministerie van Algemene Zaken Prime Minister’s Office Beleid en Maatschappij Policy and Society Beleidsanalyse Gemeenten municipal policy analysis Beleids- en Beheersinstrumentarium policy and management instruments (PMI) Besluit Begroting en Verantwoording Budget and Account Cycle Brede Heroverwegingen Comprehensive Spending Review Bureau Onderzoek en Rijksuitgaven Bureau for Research and Public Expenditures cost-benefit analysis Centraal Bureau voor Statistiek Statistics Netherlands Christen Democratisch Appel Christian Democratic Appeal Centraal Economisch Plan Central Economic Plan Christelijk Historische Unie Christian Historical Union [one of the predecessors of the CDA] Commissie voor de Ontwikkeling Committee for the van Beleidsanalyse Development of Policy Analysis Centraal Planbureau Netherlands Bureau for Economic Policy Analysis Ministerie van Cultuur, Recreatie en Ministry of Culture, Recreation Maatschappelijk Werk and Social Work [predecessor of VWS] Brede Heroverwegingen Comprehensive Spending Review Centrum voor Staatkundige Vorming Centre for Political Education [think tank of the KVP] Christen Unie Christian Union Coordinatiecommissie Coordination Committee for Wetenschappelijk Onderzoek Scientific Research into Child Kinderbescherming Protection Democraten 66 Democrats 66

List of acronyms Acronym DIV DNB ECN ECPO EGPA EIA EMU ENW EPA ERTMS EU EZ GL GR HO I&M IAK IBA IBO ICT IFFS INTOSAI IPA ISS ITS KMO KNAW KNMI

In full Dienst Informatievoorziening

English translation Department of Information Services, House of Representatives De Nederlandse Bank Dutch National Bank Energieonderzoek Centrum Energy Research Centre of the Nederland Netherlands Evaluatie en Adviescommissie Evaluation and Advisory Passend Onderwijs Committee on Special Needs Education European Group for Public Administration Environmental Impact Assessment Economische en Monetary Unie Economic and Monetary Union Expertise Netwerk voor Expertise Network for Water Waterveiligheid Safety Engineering and Policy Analysis European Railway Traffic Management System Europese Unie European Union Ministerie van Economische Zaken Ministerie van Economische Zaken Groen Links Green Left Gezondheidsraad Health Council Heroverweging van Reconsideration of Public overheidsuitgaven Expenditures Ministerie van Infrastructuur en Ministry of Infrastructure and Milieu the Environment Integraal Afwegingskader integrated assessment framework Informatiebeleid, Beleidsanalyse en Information, Policy Analysis and Automatisering Automation [unit of the VNG] Interdepartementaal interdepartmental policy Beleidsonderzoek research Information and Communication Technology Institute for Future Studies [Stockholm] International Organisation of Supreme Audit Institutions interpretive policy analysis Institute of Social Studies [The Hague] Instituut voor Toegepaste Sociologie Institute for Applied Sociology [Radboud Universiteit Nijmegen] Kleine en Middelgrote Onderneming small and medium-sized enterprise Koninklijke Nederlandse Avademie Netherlands Academy of the voor Wetenschappen Arts and Sciences Koninklijk Nederlands Royal Dutch Metereological Meteorologisch Institute Institute

xi

Policy analysis in the Netherlands Acronym KVP

In full Katholieke Volkspartij

MARIN MBO

Maritiem Research Instituut Nederland Middelbaar Beroepsonderwijs

MCA MEV MNP

Multi-Criteria Analysis Macro Economische Verkenning Milieu- en Natuurplanbureau

MoF MP NGIZ

Minister van Financiën Member of Parliament Nederlands Genootschap voor Dutch Association for Internationale Zaken International Affairs Nederlands Onderzoeksschool Netherlands Institute of Bestuurskunde Government (NIG) New Public Management Nederlandse School voor Openbaar Netherlands School of Public Bestuur Governance Nederlandse Organisatie voor Netherlands Organisation for Wetenschappelijk Onderzoek Scientific Research Organization for Economic Development and Co-operation Onderzoek en Verificatiebureau Research and Verification Bureau Public Administration Theory Network Planbureau voor de Leefomgeving Netherlands Environmental Assessment Agency Planning Programming and Budgeting System Partij van de Arbeid Labour Party Partij voor de Vrijheid Freedom Party Research and Development Rijksinstituut voor Volksgezondheid National Institute for Public en Milieu Health and the Environment Raad voor Ruimtelijk, Milieu- en Council for Spatial and Nature Natuuronderzoek Research Onderzoekscentrum voor Research Centre for Education Onderwijs en Arbeidsmarkt and the Labour Market [Maastricht University] Ruimtelijk Planbureau National Bureau for Spatial Research Rijksplanologische Dienst National Spatial Planning Agency Raad voor Werk en Inkomen Council for Work and Income Sociaal en Cultureel Planbureau Netherlands Institute for Social Research

NOB NPM NSOB NWO OECD OVB PATNET PBL PPBS PvdA PVV R&D RIVM RMNO ROA RPB RPD RWI SCP

xii

English translation Catholic People’s Party [one of the predecessors of the CDA] Maritime Institute of the Netherlands Intermediate Vocational Education Macro-economic Outlook Netherlands Environmental Assessment Agency Minister of Finance

List of acronyms Acronym SDAP SEO SEPAM SER SGP SMART SP SSK STS SZW TAW TNO UvA UWV VB VBTB VNG VSNU VVD VWS WB WBS WODC WRR

In full Sociaal-Democratische Arbeiders Partij

English translation Social Democratic Labour Party [one of the predecessors of the PvdA] Stichting voor Economisch Foundation for Economic Onderzoek Research [University of Amsterdam] Systems Engineering, Policy Analysis and Management Sociaal-Economische Raad Social and Economic Council Staatkundig Gerormeerde Partij Reformed Political Party Specific, Measurable, Attainable, Relevant and Time-bound Socialistische Partij Socialist Party Sociology of Scientific Knowledge Science and Technology Studies Ministerie van Sociale Zaken en Ministry of Social Affairs and Werkgelegenheid Employment Technische Adviescommissie voor de Technical Advisory Commission Waterkeringen for Dykes Nederlandse Organisatie voor Netherlands Organisation for Toegepast Natuurwetenschappelijk Applied Scientific Research Onderzoek Universiteit van Amsterdam University of Amsterdam Uitvoeringsinstituut Employee Insurance Agency Werknemersverzekeringen Verantwoord Begroten Accountable Budgeting Van Beleidsbegroting tot From Policy Budget to Beleidsverantwoording Accounting for Policy Vereniging Nederlandse Gemeenten; Association of Dutch Municipalities Vereniging van Samenwerkende Association of Universities in Nederlandse Universiteiten the Netherlands Volkspartij voor Vrijheid en People’s Party for Freedom and Democratie Democracy Ministerie van Volksgezondheid, Ministry of Health, Welfare and Welzijn en Sport Sport World Bank Wiardi Beckman Stichting Wiardi Beckman Foundation [think tank of the PvdA] Wetenschappelijk Onderzoek- en Scientific Research and Documentatiecentrum Documentation Centre Wetenschappelijke Raad voor het Scientific Council for Regeringsbeleid Government Policy



xiii

Policy analysis in the Netherlands

Notes on contributors Victor Bekkers is Professor of Public Administration at Erasmus University Rotterdam. He holds the chair on the empirical study of public policy and public policy processes. He is also Academic Director of the Centre of Public Innovation. Furthermore he was Dean of the Erasmus Graduate School for the Social Sciences and the Humanities. He studied Political Science and Public Administration at Nijmegen University where he has obtained his degree with honours (1987). He obtained his PhD from Tilburg University. His research interests are related to the introduction and use of new information and communication technologies as well as new media in public policy processes and the innovation challenges which emerge in their slipstream. Severine van Bommel is Assistant Professor in the Communication and Innovation Studies Group of Wageningen University. Her research and teaching are concerned with the production and use of knowledge in the fields of natural resource management and environmental governance. Her current research interests include, among others, a practice-based approach to forest and nature governance, the travel and enactment of environmental standards and the escalation of scientific controversies in relation to the legitimacy and credibility of science. She has published several international journal articles and book chapters on topics such as the (onto)politics of classifying biocultural diversity, performance of citizenship in participatory environmental governance, researchers as scientific storytellers, and the role of scientific visions in interdisciplinary and transdisciplinary environmental research. Arwin van Buuren is Associate Professor of Public Administration at the Erasmus University Rotterdam. His research is focused on questions related to water and climate governance issues and especially on the question how knowledge contributes to realising legitimate decision-making. In addition, he researches issues related to the governance of innovation in water management. Previously he has published extensively in journals such as Public Management Review, International Public Management Journal, Environmental Science and Policy, Environment and Planning C, and International Review of Administrative Sciences. In 2009 he co-edited Managing complex governance systems (Routledge) with Geert Teisman and Lasse Gerrits. He is co-editor of the recently launched journal International Journal of Water Governance (Baltzer Science Publishers). He is currently completing work on the co-edited volume Action research for climate change adaptation: Developing and applying knowledge for governance (London: Routledge). Menno Fenger is Associate Professor of Public Administration at Erasmus University Rotterdam. His research focuses on processes of long-term policy change and policy learning in social policies at the national and international level. He finished his PhD on the implementation of social policies in 2001. From 2000 to 2006 he worked as an assistant professor at the Department of Public Administration, Erasmus University Rotterdam. In 2006, he was a visiting scholar at the Center for European Studies, Harvard University. From 2006 to 2008 he was a senior policy advisor at the Dutch

xiv

Notes on contributors

Ministry of Social Affairs and Employment. In 2008, he returned to the Department of Public Administration, Erasmus University. Frank Hilterman studied Political Science at the University of Amsterdam and has a Masters in European Law and Policy from the University of Portsmouth. He worked as an assistant in the Dutch Parliament and as a financial specialist in the municipality of Haarlem. For the past 25 years he has worked for the Association of Netherlands Municipalities and was involved in the municipal policy analysis and system of budget, policy and control instruments. He was visiting lecturer at Erasmus University Rotterdam between 2010 and 2014. Peter van Hoesel has a background in Psychology and holds a PhD from Leiden University. The topic of his dissertation [1985] was the programming of policy research. He has been director of various research institutes, such as DSWO (RU Leiden), Research voor Beleid, NSS Beleidsonderzoek & Beleidsadvies (now B&A Groep), EIM and lately Panteia. He is Emeritus Professor of Applied Policy Research at Erasmus University Rotterdam. In addition he serves as chair and member of several advisory boards, member of the board and supervisory board of a couple of associations and foundations. He is doing consultancy work for the government at national, regional and local level. He is editor of the online journal Beleidsonderzoek Online. He has published widely on issues of policy research. Robert Hoppe is Full Professor of Knowledge and Policy at the University of Twente’s School for Management and Governance. He has published on topics like the governance of expertise, policy design, deliberative policy analysis and its institutional implications, the use of cultural theory in the study of public policy and technology assessment, and long-term policy dynamics. His most recent book publications are The governance of problems: Puzzling, powering and participation (Policy Press, Bristol, 2011) and (together with Hal Colebatch and Mirko Noordegraaf) Working for policy (Amsterdam and Chicago University Press, Amsterdam and Chicago, 2010). Merlijn van Hulst studied Cultural Anthropology at Utrecht University and received his PhD from Erasmus University Rotterdam. Presently, he is Associate Professor in the School of Politics and Public Administration at Tilburg University. He is interested in the role of sense-making in governance and specialises in interpretive methods. Current research includes analyses of storytelling in police practices, the work of excellent practitioners in neighbourhood governance, and the concept of framing in policy analysis. Henk Klaassen studied Spatial Economics at the Erasmus University Rotterdam (EUR). In 1979 he continued his study at the School of Architecture and Urban Planning, at the University of California Los Angeles (UCLA), USA. His thesis was on institutional and methodological aspects of policy analysis. After studying he became senior researcher at the Department of Policy Analysis, Bureau of the Budget, Ministry of Finance. In 1998 he was appointed as Associate Professor in Economics in the Department of Public Administration at the Erasmus University Rotterdam. In his research he focuses on policy analysis. Cost-benefit analysis and multi criteria analysis are central themes within that field. Other topics are New Public Management, xv

Policy analysis in the Netherlands

output steering and management control in local governments. He in editor-in-chief of TPC, a Dutch magazine on public governance, audit and control. Peter van der Knaap is Managing-Director of SWOV Institute for Road Safety Research in The Hague, the national institute for scientific road safety research in the Netherlands. Prior to this he was Performance Audit Director with the Netherlands Court of Audit, which aims to audit and improve the regularity, efficiency, effectiveness and integrity with which the state and associated bodies operate. He also held the position of Head of Policy Evaluation and Analysis at the Ministry of Finance,The Hague, and before that he was a researcher at Erasmus University Rotterdam. Van der Knaap earned his PhD at the same university in 1997. He publishes on policy evaluation, outcome budgeting and policy-oriented learning and is a member of the editorial board of the journal Evaluation. Pierre Koning has been Chief Science Officer at the Ministry of Social Affairs and Employment since July 2011, and part-time Professor on Labour Economics at the VU University of Amsterdam. Pierre was previously employed as a Program Leader on Public Organisation at CPB Netherlands Bureau for Economic Policy Analysis. Pierre obtained his PhD at theVU University of Amsterdam – with his thesis focusing on job search models for the labour market. Pierre’s special interest lies in the field of welfare-to-work services and disability insurance – also the focus of most of his international publications. In 2009, Pierre was a visiting scholar at the Institute of Research on Poverty at the University of Wisconsin-Madison. Joop Koppenjan is a Professor of Public Administration at the Erasmus University Rotterdam. His research focuses on public policy, network governance, public private partnerships and public service delivery. Areas of application are infrastructure-based sectors such as transport, water, and energy, and social sectors such as social support, care, and safety. This research often has a comparative focus, either between policy areas or between countries. Recently he published The new public governance in public service delivery (The Hague: Eleven, 2012). Frans van Nispen holds a MPA from Leiden University and a PhD from Erasmus University Rotterdam. He served for several years as a policy analyst for the Dutch government, before he returned to academia. He has been Affiliated Professor at the Institute of Public Policy at George Mason University, Fairfax, Virginia, and Senior Research Fellow at the Robert Schuman Centre for Advanced Studies at the European University Institute in San Domenico di Fiesole, Italy. At the moment he is Associate Professor of Public Administration at the Erasmus University Rotterdam. He has published in various journals, primarily on issues at the interface of policy analysis and public budgeting in a European context. He has done consultancy for, inter alia, the EU, the OECD and the World Bank. Edwin van Rooyen is a researcher at the Montesquieu Institute and the Documentation Centre for Dutch Political Parties. He has written a dissertation on the Europeanisation of interest groups and political parties in the Netherlands. His current research focuses on the interaction between Dutch and European political

xvi

Notes on contributors

parties. His research interests include European integration, interest representation and decision-making in the EU. Rien Rouw is senior policy advisor at the Dutch Ministry of Education. Previously he worked at the General Court of Audit and the Council for Social Development in the Netherlands. He published several articles in Dutch journals on diverse themes such as violence, values and standards, education, healthcare, media and the relationship between government and society. Peter Scholten is Associate Professor of Public Policy & Politics at the Erasmus University of Rotterdam. He is also an associate researcher at the Centre on Migration, Policy and Society at the University of Oxford. At Erasmus University, Peter coordinates a research cluster on ‘The governance of migration and integration in Europe’. He is currently involved in various international research projects, such as on the multi-level governance of migrant integration, on the migration-integration nexus and on science-society dialogues on migrant integration in Europe. Peter has published in various international journals, and recently published his book Framing immigrant integration: Dutch research-policy dialogues in comparative perspective (Amsterdam University Press, 2011). In addition, he is one of the founding editors of the new International Journal of Comparative Migration Studies, which was published for the first time in 2013, and editor of Perspectives on Europe, the journal of the Council for European Studies. Peter is a Fellow of the Montesquieu Institute for European Studies, coordinator of the IMISCOE research cluster on ‘Research-policy dialogues on migration and integration in Europe’, a member of the Immigration Research Group of the Council for European Studies (CES), a senior member of the Netherlands Institute for Government (NIG) and a member of the board of the Dutch Centre for History of Migrants. Martijn van der Steen is Associate-Dean and Deputy-Director of the Netherlands School of Public Administration (NSOB) in The Hague, the Netherlands. His research focuses on public sector strategy, long-term policy making, and governance in networks. He holds a Masters in Public Administration (Erasmus University Rotterdam, EUR) and a Masters in History (EUR). In 2009 he received his Doctorate at Tilburg University.  Wil Thissen is a Professor of Policy Analysis and head of the Policy Analysis Section in the Faculty of Technology, Policy and Management at Delft University of Technology (TU Delft). After graduating in Applied Physics, he received a PhD in Systems and Control Engineering from Eindhoven University of Technology, and was affiliated with the University of Virginia and the Dutch Public Works Department before joining TU Delft in the mid-1980s. He was one of the pioneers in the development and realisation of the teaching and research programme in Systems Engineering, Policy Analysis and Management at TU Delft, and director of the Multi-Actor Systems research programme. His current research interests are in developing and testing concepts and methods for supporting strategic policy making in multi-actor environments, with particular emphasis on applications in the field of infrastructures, energy policy, and environmental and water management. He served as an editorial

xvii

Policy analysis in the Netherlands

board member of Technological Forecasting and Social Change, Impact Assessment and Project Appraisal, and The Environmental Impact Assessment Review. Arco Timmermans is Research Director of the Montesquieu Institute in The Hague and Associate Professor of Comparative Public Policy at the Institute of Public Administration, Leiden University. He obtained his PhD in the social and political sciences at the European University Institute in Florence. Previously he worked at the Departments of Public Administration of the Delft University of Technology and the University of Twente, and was a guest lecturer at several universities across Europe. His research focuses on attention to policy problems and agenda setting in the Netherlands in comparative perspective and in the European Union. Projects on this subject analyse the attention to problems in government plans since 1945; links between media attention, scientific information and political priority for environmental issues such as climate change; the expansion of attention to safety issues; budgetary changes over time; cross-country comparisons of immigrant policy and morality policy, and the impact of Europeanisation on the national policy agenda. He also publishes on coalition governance and legislative politics in comparative perspective. Mark van Twist is Professor of Public Administration at Erasmus University Rotterdam. He is also Dean and member of the board of the Netherlands School of Public Administration, an inter-university institute in The Hague that offers study programmes for senior officials in government, and extraordinary member of the board of the Netherlands Court of Audit. He has published numerous contributions to books and articles, eg in the International Review of Administrative Sciences, Public Integrity and Public Management Review. Gerrit Voerman is a historian and holds the chair in Development and Functioning of the Dutch and European Party Systems at the University of Groningen. He studied History at the University of Groningen and gained his PhD in 2001 with a thesis entitled ‘The Moscow meridian: The CPN and the Communist International (1919–1930)’. Voerman has headed the Documentation Centre for Dutch Political Parties (DNPP) of the University of Groningen since 1989. He publishes extensively on political parties. He is the editor of a book series on the political parties represented in the Dutch House of Representatives. He is currently researching populism, party organisation, party identity in an international perspective, and the relationship between Dutch parties and the European parties of which they form part. Dvora Yanow is a policy and organisational ethnographer and interpretive methodologist whose research and teaching are shaped by an overall interest in the communication of meaning in organisational and policy settings. She is Guest Professor in the Communication, Philosophy, and Technology sub-department at Wageningen University (NL) and Professor of Organisational Studies at Keele University (UK). Her policy-focused research investigates state-created categories for race-ethnic identity, immigrant integration policies and citizen-making practices, policy frames and framing, and research regulation policies and practices (ethics review committees. Her most recent book is Interpretive research design: Concepts and processes (Routledge), co-authored with Peregrine Schwartz-Shea, with whom she co-edits the Routledge Series on Interpretive Methods. Other current projects engage science/technology xviii

Notes on contributors

museums and the idea of science, practice theory and the life cycle, and space analysis methods. Karin Zaal is Head of the Parliamentary Bureau for Research and Public Expenditure of the House of Representatives of the States General. She started there in August 2010. Being trained as a social psychologist at the University of Amsterdam, she worked for about eight years as a senior scientific researcher at TNO Healthy for Life, mainly on subjects such as ‘ageing well’ and living arrangements for the elderly. After that she worked for the Provincial Government of North-Holland for 14 years in research and management positions and for the Ministry of Education, Culture and Science as a senior operational auditor for two years.

xix

Policy analysis in the Netherlands

Editors’ introduction to the series Professor Iris Geva-May and Professor Michael Howlett, ILPA series editors

Policy analysis is a relatively new area of social scientific inquiry, owing its origins to developments in the US in the early 1960s. Its main rationale is systematic, evidencebased, transparent, efficient, and implementable policymaking. This component of policymaking is deemed key in democratic structures allowing for accountable public policies. From the US, policy analysis has spread to other countries, notably in Europe in the 1980s and 1990s and in Asia in the 1990s and 2000s. It has taken, respectively one to two more decades for programmes of public policy to be established in these regions preparing cadres for policy analysis as a profession. However, this movement has been accompanied by variations in the kinds of analysis undertaken as USinspired analytical and evaluative techniques have been adapted to local traditions and circumstances, and new techniques shaped in these settings. In the late 1990s this led to the development of the field of comparative policy analysis, pioneered by Iris Geva-May, who initiated and founded the Journal of Comparative Policy Analysis, and whose mission has been advanced with the support of editorial board members such as Laurence E. Lynn Jr., first co-editor, Peter deLeon, Duncan McRae, David Weimer, Beryl Radin, Frans van Nispen, Yukio Adachi, Claudia Scott, Allan Maslove and others in the US and elsewhere. While current studies have underlined differences and similarities in national approaches to policy analysis, the different national regimes which have developed over the past two to three decades have not been thoroughly explored and systematically evaluated in their entirety, examining both sub-national and non-executive governmental organisations as well as the non-governmental sector; nor have these prior studies allowed for either a longitudinal or a latitudinal comparison of similar policy analysis perceptions, applications, and themes across countries and time periods. The International Library for Policy Analysis (ILPA) series fills this gap in the literature and empirics of the subject. It features edited volumes created by experts in each country, which inventory and analyse their respective policy analysis systems. To a certain extent the series replicates the template of Policy Analysis in Canada edited by Dobuzinskis, Howlett and Laycock (Toronto: University of Toronto Press, 2007). Each ILPA volume surveys the state of the art of policy analysis in governmental and non-governmental organisations in each country using the common template derived from the Canadian collection in order to provide for each volume in the series comparability in terms of coverage and approach. Each volume addresses questions such as: What do policy analysts do? What techniques and approaches do they use? What is their influence on policy-making in that country? Is there a policy analysis deficit? What norms and values guide the work done by policy analysts working in different institutional settings? Contributors focus on the sociology of policy analysis, demonstrating how analysts working in different organisations tend to have different interests and to utilise different techniques. The central theme of each volume includes historical works on the origins of policy analysis in the jurisdiction concerned, and then proceeds to investigate the nature and types, xx

Editors’ introduction to the series

and quality, of policy analysis conducted by governments (including different levels and orders of government). It then moves on to examine the nature and kinds of policy analytical work and practices found in non-governmental actors such as think tanks, interest groups, business, labour, media, political parties, non-profits and others. Each volume in the series aims to compare and analyse the significance of the different styles and approaches found in each country and organisation studied, and to understand the impact these differences have on the policy process. Together, the volumes included in the ILPA series serve to provide the basic data and empirical case studies required for an international dialogue in the area of policy analysis, and an eye-opener on the nuances of policy analysis applications and implications in national and international jurisdictions. Each volume in the series is leading edge and has the promise to dominate its field and the textbook market for policy analysis in the country concerned, as well as being of broad comparative interest to markets in other countries. The ILPA is published in association with the International Comparative Policy Analysis Forum, and the Journal of Comparative Policy Analysis, whose mission is to advance international comparative policy analytic studies.The editors of each volume are leading members of this network and are the best-known scholars in each respective country, as are the authors contributing to each volume in their particular domain. The book series as a whole provides learning insights for instruction and for further research in the area and constitutes a major addition to research and pedagogy in the field of comparative policy analysis and policy studies in general. We welcome to the ILPA series Volume 3, Policy Analysis in the Netherlands, edited by Frans van Nispen and Peter Scholten, and thank the editors and the authors for their outstanding contribution to this important encyclopedic database. Iris Geva-May Professor of Policy Studies, Baruch College at the City University of New York, Professor Emerita Simon Fraser University; Founding President and Editor-in-chief, International Comparative Policy Forum and Journal of Comparative Policy Analysis Michael Howlett Burnaby Mountain Professor, Department of Political Science, Simon Fraser University, and Yong Pung How Chair Professor, Lee Kuan Yew School of Public Policy, National University of Singapore

xxi

Policy analysis in the Netherlands

Preface and acknowledgements Policy analysis is a long-established field in the Netherlands, in academia as well as in practice. It makes the Netherlands one of the strongholds of policy analysis, one that certainly deserves attention in a special volume entirely devoted to the evolution of policy analysis. As the Dutch government adopted an Anglo-Saxon stance immediately after World War II, generations of students in Political Science and Public Administration have been educated and trained with American textbooks and are thus familiar with the essentials of policy analysis and, being in leading positions now, equipped to deal with complex and wicked problems in society. The volume brings together contributions from key Dutch scholars as well as from practitioners working in the field, crossing the layers of government as well as the borders between policy domains. We would like to thank them all for sharing their insights and observations with a broader audience. We believe that the outcome is more than just a collection of loosely coupled articles. We hope that this volume will serve as a benchmark for countries that are just starting up and will induce a dialogue about the added value of policy analysis, especially now we are facing the far-reaching consequences of the sovereign debt crisis. We are grateful to the series editors of the International Library of Policy Analysis – Iris Geva-May and Michael Howlett – for offering a platform for this book and for their guiding scholarship that has been an important inspiration. In addition, we would like to express our appreciation to the team at Policy Press for their dedication, patience and support. We are furthermore heavily indebted to Yneke Steegstra for helping us to meet the deadlines and the guidelines set by the publisher. Last but not least, we owe thanks to the board of the Comparative Public Services Innovation research group at the Erasmus University Rotterdam for its grant, which enabled us to deal with setbacks during the process. Frans van Nispen and Peter Scholten Rotterdam, 1 August 2014

xxii

ONE

Policy analysis in the Netherlands: an introduction Frans van Nispen and Peter Scholten

1.1: Introduction The Netherlands is commonly regarded as one of the strongholds of policy analysis, both in academia and in policymaking. Few countries have such a high density of institutes specialised in policy analysis, and in few countries have such institutes played and continue to play such a key role in policymaking as in the Netherlands. For instance, so-called planning bureaus, a specific Dutch phenomenon, have a played a central role not just in the evaluation and monitoring of policies, but also in setting issues on the political agenda, raising policy alternatives and sometimes even directly affecting political decision-making. Even in the political arena, it is institutionalised practice in the Netherlands that policy analyses are made by authoritative institutes of the social and economic impacts of the political programmes of different parties. Policy analysis is clearly an authoritative factor without which it is impossible to comprehend the dynamics of policymaking and politics in the Netherlands. This is also reflected in the strong presence of policy analysis in research groups and teaching at universities throughout the Netherlands. Most Dutch universities have research groups specialised in policy analysis, sometimes as part of departments of public administration, but sometimes also as part of more specialised departments for the technological or agricultural sector. Some of these groups have played a key role in the development of the international literature on policy analysis, for instance, providing a stronghold for the development of rational as well as critical and postempiricist perspectives on policy analysis. Moreover, in between universities and policy agencies, an amalgam of institutes has evolved that are specialised in policy analysis.The Dutch setting is characterised by an abundance of what can be described as ‘boundary organisations’ (Halffman and Hoppe, 2005). The aim of this book is to provide an overview of developments in policy analysis in terms of academic thinking, as well as in terms of its role in policy and politics in the Netherlands. It brings together contributions from key Dutch scholars in this field, as well as from practitioners from institutes specialised in policy analysis. Rather than focusing on one of the schools of academic thinking or practices of policy analysis, we attempt to capture the diversity of academic thinking and policy analysis practices as evolved in the Netherlands over the last decades. We believe that it is this diversity and plurality of perspectives that positions the Dutch case in an international comparative setting, as well as accounts for the strength of policy analysis in Dutch academia and policymaking.Various contributions in this volume explicitly pay attention to this diversity in perspectives on policy analysis. Furthermore, 1

Policy analysis in the Netherlands

in each of the contributions, we will substantiate empirically the role that policy analysis can play in the Netherlands by looking at cases not just from various types of institutes, but also from different policy domains.This includes such diverse domains as financial policies, public management, education policies, welfare state policies, water governance and migrant policies. In this introduction, we first sketch the contours of Dutch policy analysis, in relation to academia as well as to government (Dunn, 1983: 1), in an effort to position the Dutch case for an international audience. Then, we will provide a more conceptual analysis of what is referred to as policy analysis, followed by an outline of the different sections and contributions of the book. Finally, we sketch the contours of what we believe to be a valuable agenda for the future of policy analysis in the Netherlands and beyond.

1.2: Policy analysis in Dutch academia Policy analysis in the Netherlands has been driven by specific developments in Dutch academia. In the period of genesis of many of the Dutch institutes for policy analysis, the 1970s, there was already a strong policy orientation among the social sciences in the Netherlands (Blume et al, 1991). There was a strong willingness to connect knowledge and policymaking, and personal networks between research and policymakers were particularly strong. This policy orientation has persisted ever since among at least a significant proportion of Dutch social scientists, while there has undeniably also been a significant growth of a more academic (and international) orientation in Dutch policy analysis. Most Dutch universities have established research groups specialised in policy analysis, and many provide specific teaching programmes on policy analysis. In most large universities, policy analysis is incorporated as an integral part of public administration departments and teaching in public administration and policy sciences (eg Leiden University, the University of Amsterdam, the Free University of Amsterdam, Erasmus University Rotterdam, University of Utrecht,Tilburg University, Nijmegen University, Twente University and Maastricht University). Several universities, such as Delft University of Technology and Wageningen University, provide more specialised programmes in policy analysis for the technological sector and for the agricultural sector, respectively. In addition, several inter-university institutes have evolved in which policy analysis plays a key role, like the Netherlands School for Public Governance (NSOB) and the Netherlands Institute of Government (NIG). The proliferation of policy analysis in academia over the past decades has also contributed to a growing diversity in theoretical perspectives on policy analysis. This can be considered a signal of maturity of policy analysis as an academic field in the Netherlands. At the same time, it has also contributed to what Goemans (1988: 351) has described as a ‘battle of analysis’ or Radin (2000: 31) as ‘dueling swords’: government is often faced with the impossible task of encompassing various and sometimes contradicting studies under one umbrella (Klaassen and Van Nispen, 1996). In the international literature on policy analysis, Dutch scholars have played an important role in the development of very different schools of thinking about policy analysis. In this book, we bring together not only contributions that cover various aspects of policy analysis and evaluation in the Netherlands, but also contributions by Dutch scholars that represent different ‘schools of thinking’ about policy analysis in 2

Policy analysis in the Netherlands: an introduction

the Netherlands. In fact, many of the Dutch scholars represented in this book have also played a key role in the international development of these different schools of thought in policy analysis. First of all, as Bekkers shows in this volume (Chapter Fifteen), in the 1980s, academic thinking about policy analysis in the Netherlands was already characterised by plurality, with rational as well as more political perspectives. In particular, Hoogerwerf has been internationally recognised as one of the founding fathers of the rational perspective of policy analysis, whereas other Dutch scholars like Snellen, Ringeling and later Hoppe have been recognised internationally for their contribution to a more critical and political perspective. However, as various contributions to this volume show, diversity has increased, in particular, since the 1990s. One of the perspectives to which Dutch scholars have contributed in particular involves the network perspective, which involves analysis of policies in complex networks of actors (elaborating many elements of the earlier political perspective). Koppenjan and Van Buuren, both important representatives of this mode of policy analysis, show in this volume how network analysis offers not only a different perspective on policy analysis, but also a different approach to policy analysis, involving processes of active joint fact-finding of the policy analyst together with policy stakeholders. Another strand of policy analysis in the Netherlands elaborates more on the institutionalist perspective of policy analysis (such as March and Olsen, 1989). Of particular relevance in the Dutch context and beyond has been the work of Hemerijck (2001, 2003), who elaborated on March and Olsen’s distinction between the logic of consequence and the logic of appropriateness in four questions about sources of legitimacy of policies. In this volume, Fenger and Koning, among others, elaborate on this type of policy analysis when referring to the relevance of ‘negotiated knowledge’.This type of policy analysis is closely associated with more participatory and interactive forms of policy analysis, as advocated by various Dutch scholars (see Bekkers, Chapter Fifteen, and Hoppe, Chapter Four). Finally, perhaps one of the strands of policy analysis on which Dutch scholars have left their clearest mark is the so-called social-constructivist approach to policy analysis, which is composed of several sub-strands. First, it includes what Fischer and Forrester (1993) have described as the argumentative turn in policy analysis, which articulates the importance of public debate and free speech. Second, a more cultural or interpretive policy analysis that attributes particular attention to the role of deliberation in policy-making and -implementation (Hoppe, 1998). As Van Bommel, Van Hulst and Yanow show in this volume, Dutch scholars have been strongly involved in the development of interpretive policy analysis from the very beginning, somewhere in the 1980s. Dutch scholars like Wagenaar, Hajer and Hoppe, as well as Yanow, who is currently based in the Netherlands, closely worked together with international scholars like Rein, Fischer and Laws in pioneering this mode of policy analysis. Since then, Dutch scholars and research groups at specific Dutch universities (like Tilburg University, Leiden University, Amsterdam University and Twente University) have continued to play a key role in this perspective. However, as Van Bommel et al show in this volume, it does not make sense to speak of one coherent ‘paradigm’ of interpretive policy analysis in the Netherlands, as different sub-strands have emerged over the past decade or so (including discourse analysis, framing analysis, category analysis, narrative analysis and practice studies). 3

Policy analysis in the Netherlands

1.3: Policy analysis in the Dutch polity Besides these more academic factors, we can also point at several policy or political factors that have influenced the development of policy analysis in the Netherlands. As various contributions in this volume show, the Dutch case is characterised by a high density of institutes specialised in policy analysis and a strongly institutionalised role of policy analysis in policymaking.Various factors can be identified that account for this (in comparative perspective) relatively strong role of policy analysis in the Netherlands. Perhaps most commonly referred to is the consociational character of Dutch politics and democracy (Lijphart, 1969). Research and expertise have traditionally played a key role in consensus-building in the highly fragmented political system in the Netherlands (see also Blume et al, 1991). Expertise provided a non-partisan and authoritative source on which political consensus could be constructed, and may also have helped to avert political controversy on policy issues that could threaten the political consensus (see also Scholten, 2011). Even now that the consociational character of Dutch politics appears to be weakening, there is a continued demand for ‘evidence-based policymaking’, not just in government departments, but increasingly also in politics, such as with the establishment of parliamentary investigative and inquiry committees. The consociational character of Dutch politics also helps account for three specific traits of policy analysis in the Netherlands. First, it has enabled the founding of relatively independent institutes for policy analysis. It is this relative independence that provides these institutes with the authority needed to forge consensus on a variety of policy topics. For instance, the Netherlands is one of very few countries to have established a high-level scientific council to advise government and politics, the Scientific Council for Government Policy (Wetenschappelijke Raad voor het Regeringsbeleid; WRR); in many other countries, the establishment of such a council would have been considered an obstruction of political primacy (Scholten, 2011). The same applies to the Council of State, a high council established already in the 16th century with a strongly institutionalised role in advising government on new policies and legislation. A second specific trait of policy analysis in the Netherlands that the consociational setting helps account for involves the mix of ‘interests-cum-expertise’ that continues to play a key role in Dutch politics up to this very day. As an exponent of the Dutch ‘polder model’, involving cooperation of employer organisations, trade unions and government in addressing a variety of social and economic issues, the use of tripartite structures for policy analysis and policy advice has been common practice in the Dutch setting.The most important example in this respect involves the Social and Economic Council (Sociaal-Economische Raad; SER).The SER is the most important advisory board for the cabinet and Parliament on social-economic issues. It brings together representatives from employer organisations and employee organisations, as well as independent experts (so-called ‘crown members’). A third trait of policy analysis in the Netherlands that consociationalism accounts for is the relatively weak politicisation of expertise.The politicisation of expertise and the role of party-political think tanks has been much less developed in the Netherlands than, for instance, in the UK or the US (Fischer, 1993).Whereas most political parties have established, or are associated with, specific party-political think tanks, their role in Dutch politics has remained relatively marginal, especially when compared with more independent advisory bodies and planning bureaus. More generally, the Dutch 4

Policy analysis in the Netherlands: an introduction

system is characterised by an asymmetry in terms of the information and knowledge position of political parties and Parliament versus that of government departments. However, the last decade especially has witnessed a gradual strengthening of the information and knowledge position of Parliament in particular. Among others, it has established a Parliamentary Bureau for Research and Public Expenditure (Bureau Onderzoek en Rijksuitgaven; BOR) to facilitate the growing number of parliamentary investigative or inquiry committees that have been established over the last decade. It is important to recognise the diversity of the institutional landscape of policy analysis in the Netherlands as it is today. First, some forms of policy analysis are incorporated into government agencies. For instance, like Van Nispen shows in this volume, already in the 1970s, government established a high-level inter-ministerial Committee for the Development of Policy Analysis (Commissie voor de Ontwikkeling van Beleidsanalyse; COBA) to promote the institutionalisation of policy analysis within government departments and thus to promote the accountability of policies. Second, there is an abundance of government-associated but, in practice, relatively independent institutes specialised in policy analysis. This includes the so-called planning bureaus that are often associated with specific government departments but are allowed to operate relatively independently. Planning bureaus like the Social and Cultural Planning Office (Social en Cultureel Planbureau; SCP), the Central Planning Bureau (Centraal Planbureau; CPB) and the Environmental Assessment Agency (Planbureau voor de Leefomgeving; PBL) are forces to be reckoned with when accounting for policymaking in the Netherlands. Third, there are a number of institutes with a very special position and strong authority, which are often very specific to the Dutch case.This includes the Council of State, the SER and the Scientific Council for Government Policy. Like some of the planning bureaus, these special institutes are clear examples of how policy analysis in the Netherlands plays a role well beyond that of monitoring and evaluating public policies; they can also be powerful agenda setters and sources of policy ideas, and can even play a key role in politics. The fact that Dutch media usually find their way to these Dutch institutes very easily has provided them with even more authority and more diverse ways of affecting policy. Finally, there has also been a proliferation of a commercial market for policy analysis. Various commercial institutes for policy analysis have come up over the last decade or so, such as Ecorys and Panteia.

1.4: A conceptual map of policy analysis in the Netherlands and beyond From the start, policy studies consisted of two varieties: an academic variant that focused on explanations of causes and consequences of a policy-focused variant; and an administrative variant that was concerned with the improvement of policy (Lasswell, 1951; Dunn, 1983).1 The latter is also referred to as policy research or policy studies (Howlett et al, 2009: 8). The difference may be summarised as in Table 1.1. In this contribution, we limit ourselves to the latter variety of policy analysis, which includes policy evaluation.2 Following Dunn (1994: 62), we define policy analysis as applied social science research (in a broad sense) focused on the production of arguments and information that can be used in the political arena. This definition 5

Policy analysis in the Netherlands

Table 1.1: Policy analysis versus policy research or studies Policy studies

Policy analysis

Focus

Theory-driven

Utilisation-focused

Scope

Nomological

Ideographic

Discipline

Mono-disciplinary

Multidisciplinary

Approach

Descriptive

Prescriptive

Type of knowledge

Empirical

Normative

presupposes that complementary information results in better policy. The question is not so much whether policy works (explanation), but whether, and if yes, how, the influence of policy can be improved (manipulation).3 It is focused on goal achievement, which needs to be carefully distinguished from effectiveness and efficiency. In practice, efficiency must be weighed continuously against an equitable and fair distribution of goods and services. The domain of policy analysis gradually expanded to political feasibility and societal acceptability of policy when it became clear that the results of research were often not used because of a lack of societal support. Contrary to operations research and systems analysis, both of which stood at the cradle of policy analysis (Majone, 1985: 35–49), the results of policy analysis are not judged on the basis of economic criteria such as efficiency, but on non-economic, i.e. political, criteria such as feasibility.Thus, policy analysis can be regarded as ‘systems analysis writ large, in the sense that it includes, in addition to the technical and economic aspects of a policy problem, also the political aspects that systems analysis is supposed to have overlooked’ (Majone, 1985: 47). It is no longer the case that a policy analyst can suffice with a report that sketches alternatives in terms of their (dis)advantages.Thus, Majone (1975: 50) observed that: ‘feasibility, rather than optimality, is a realistic goal for policy analysis’. He defines feasibility in terms of limitations to be distinguished between administrative, economic, institutional, political, social and technical conditions. In the Netherlands, Van de Vall (1987) especially called attention to the criteria for judging the results of policy-focused research. He distinguished between scientific (epistemological) and non-scientific (implementation or strategic) parameters or validity. A policymaker is not waiting to see a large number of explanatory variables, of which none appears to be manipulable, but will also consider other conditions. There is no sense in proposing something for which there is no support in politics or society (Leeuw, 1987: 164). The results of research that is too late are perceived as mustard after dinner (Knott and Wildavsky, 1980: 548).A proposal that violates the law stands little chance of success unless a parliamentary majority can be found to change the law (Patton and Sawicki, 1986: 277). One of the more recent developments in policy analysis, which as mentioned has been particularly manifest in the Netherlands, concerns the so-called ‘argumentative turn’ in policy analysis.The deliberative mode of policy analysis attributes particular attention to the role of arguments, language and interaction in policies. For policy analysis, it has brought about an important shift in focus from the policies themselves and their outputs/outcomes, to the process of policy-making and -implementation. It has driven policy analysts to increasingly study the interaction between policy actors or stakeholders as an integral part of policy analysis. 6

Policy analysis in the Netherlands: an introduction

Implementation analysis According to Potman (1989), policy design ought to focus more on the acceptance or implementation (Korsten, 1985; Maarse, 1986; Hoppe et al, 1987) of policy, and as a result, focus more on possible barriers and how these can be removed.‘In conscious anticipation of Murphy’s Law, advocates might continually ask,“What can go wrong?” as they design policies and set up mechanisms for program delivery’ (Grindle, 1980, quoted in O’Toole, 1986: 193). One of the methods of finding possible bottlenecks that may occur during implementation – ‘an exercise in concentrated pessimism’ (Bardach, 1980: 255) – is to develop implementation scenarios where a distinction can be made between forward- and backward-mapping (Elmore, 1979). In terms of Table 1.1, implementation analysis concerns the political feasibility (Does it fit?) and societal acceptability (Is it appropriate?) of policy. Following Snellen (1987: 4), we can add the constitutional or legal foundation (Is it allowed?), which includes attention to equality before the law, due process and freedom from capriciousness. Finally, it is important to observe that none of the criteria are dominant. In practice, an assessment must often be made between various criteria since societal problems do not stop at disciplinary boundaries. In recent years, the ‘trade-off ’ often went to the pursuit of efficiency but the pendulum appears to be swinging back.The government is increasingly moving away from deregulation, privatisation and contracting out.4

Political feasibility According to Meltsner (1972: 859, in reference to Dror, 1967: 200), a policy analyst should not only pay attention to technical feasibility, but also to political feasibility, in order to prevent the research results from disappearing to the bottom of the pile or even ending up in the rubbish bin. An analysis of political feasibility includes an inventory of the situation in terms of ‘the actors involved, their beliefs and motivations, the resources they hold, their effectiveness in using these resources and the sites at which decision will be made’ (Patton and Sawicki, 1986: 285). In developing and comparing solutions, the analysis can be expanded to ‘possible political problems, potential supporters and opponents, areas of compromise, new alternatives that might be considered, existing alternatives that might be modified to gain supporters, and steps to take to enhance opportunities for implementation’ (Patton and Sawicki, 1986: 285). In practice, attention needs to be given first to the responsible minister who has placed the item on the political agenda: how hard is he or she willing to defend the proposal, what does the power network look like, and how is the lay of the votes in Parliament? According to Majone, policy analysis can be elaborated in two ways: 1. an analysis of political feasibility after the selection of the desired alternative; and 2. the analysis of political feasibility during each phase of the investigation. The last possibility is, obviously, preferable, but also more difficult to achieve since it requires, in the words of Dror (1984: 116; see also Radin, 2000: 100), a ‘paradigmatic innovation’ in policy analysis.

7

Policy analysis in the Netherlands

Societal acceptability Closer inspection of the concept of political feasibility shows that it needs to be interpreted broadly and include societal acceptability, which is support or foundation in society for the implementation of a proposal. Following Wilson (1980), a distinction can be made between the degree to which the benefits and the costs of the intended policy are disseminated across the entire population, which results in four options. In the first case, (dis)advantages are concentrated and institutionalised in an interest group. The relative power position is decisive for the policy (interest policy). The situation is relatively simple when the benefits meet a target group but the costs are spread over the entire population. In this case, a policymaker is guided by the interests of the target group (client policy). In the reverse situation where part of the population pays for the costs while the entire population enjoys the benefits, someone is often needed to break through the power barrier (entrepreneurial policy).5 The democratic ideal of decision-making by means of majority rule applies to the situation where (dis) advantages are more or less equally divided across the population (majoritarian policy).

Table 1.2: A classification of public policy Costs

Concentrated

Distributed

Concentrated

Interest group policy

Client policy

Distributed

Entrepreneurial policy

Majoritarian policy

Benefits

Source: Wilson (1980)

Two remarks that will nuance this are in order. First, we observe that Table 1.2 is no more than a heuristic tool, which contains insufficient variety. Using such a schema, a policymaker can determine what to consider. Second, the schema can be applied to the existing, as well as to a desired, situation. These do not necessarily coincide. A policymaker can find a hold in the group that will benefit from the intended policy, or with the group that is confronted with the costs of the current policy.

Legal foundation Finally, a proposal needs to be checked for its legality or due process. A measure violating the law can expect little acceptance unless there is a majority to change the law. That, however, takes time, which may be an insurmountable barrier. In this case, a suboptimal result that fits into the existing constitutional or juridical framework must be accepted.Testing for due process is thus a significant component of policy analysis in the view of the (summary) scorecard in the Oosterschelde study.6 The alternatives proposed are assessed with the question of whether they fit the so-called Delta Act, among others.The developments in the field indicate a constantly expanding domain of policy analysis (see Figure 1.1).

8

Policy analysis in the Netherlands: an introduction

Figure 1.1: A framework for policy analysis Legitimacy

Output legitimacy

Input legitimacy

Logic of appropriateness

Feasibility

Acceptability

Logic of consequence

Effectiveness

Legality

Logic

Source: Hemerijck and Van Nispen (2006)

A last point has to do with the relation between policy analysis and policy evaluation. Following Dunn (2004), a distinction could be made between prospective and retrospective policy analysis. We reserve the term ‘evaluation’ for the latter, although it is sometimes used for the appraisal of a policy. The evaluation may be further refined by making a distinction between theory-driven evaluation, which is mostly carried out in academia, and utilisation-focused evaluation (Patton, 2008), which is practised by governments. The interest in evaluation got a boost by the New Public Management (NPM) movement and the subsequent introduction of performance budgeting. The NPM seems to be over the hill (Dunleavy et al, 2005; Kraan, 2010), making policymakers look at alternative sources of information to deal with the consequences of the sovereign debt crisis. A promising route is provided by spending reviews, which may be considered as a mode of programme evaluation in times of austerity (Van Nispen, 2014). A spending review may be considered as a form of integrated policy analysis, that is, a combination of prospective and retrospective policy analysis, or, more precisely, policy evaluation. On the one hand, spending reviews look backwards as they are geared to an assessment of timeliness, effectiveness and/or efficiency. On the other hand, spending reviews look forwards as they are supposed to generate better options for the future and may be characterised as utilisation-focused evaluation. Contrary to performance evaluations, spending reviews are focused on the generation of smart or targeted cuts and, as such, compensate for ‘the fundamental asymmetry of the regular budget process which is capable of producing good options for new spending, but not of producing good options for new savings’ (OECD, 2011: 81).

1.5: Outline of this book This book consists of five parts that focus on different dimensions of policy analysis in the Netherlands. Part One focuses on styles and methods of policy analysis. This part tries to capture the diversity in styles and methods that, as argued earlier, we see as characteristic for policy analysis in the Netherlands. It also tries to connect these styles and methods to more generic developments in the literature on policy analysis, thus positioning the Dutch case in a comparative perspective. In their contribution (Chapter Two), Van Twist, Rouw and Van der Steen provide an original analysis of how the concept of ‘evidence-based policy analysis’, derived from the rational approach to policy analysis, applies to the Dutch case. In particular, they look at how evidence-based policy analysis is institutionalised within knowledge units of 9

Policy analysis in the Netherlands

government organisations at different levels of government. Subsequently, in Chapter Three, Van Buuren and Koppenjan describe the relevance of the network approach in the Dutch setting in particular, also applying the network approach to a typical Dutch case study (water governance in a Dutch polder). In his contribution on ‘Patterns of science–policy interaction’ (Chapter Four), Hoppe provides an analysis of different forms of interaction between research and policy in the Dutch setting, distinguishing between neo-corporatist, neoliberal and more deliberative patterns of interaction. Finally, in Chapter Five, Van Bommel, Van Hulst and Yanow approach the Dutch case from a fourth theoretical perspective in which Dutch scholars have played a particular powerful role: interpretive policy analysis. They not only provide a unique insight into the role of Dutch scholars in the genesis of this approach, but also convincingly show how the institutionalisation of this approach in the Dutch context also led to growing diversity within this approach. Part Two zooms in on policy analysis by government organisations themselves, both within government agencies and as an institutional practice of these agencies. In his contribution, Van Nispen (Chapter Six) traces the institutionalisation of policy analysis within national government agencies back to its very beginning in the 1970s. Subsequently, he shows how the system of ‘From Policy Budget to Accounting for Policy’ became part of everyday practice of government agencies up to this very day. Strongly related to the developments that Van Nispen signals for the national level, Hilterman and Klaassen (Chapter Seven) show in their contribution that similar processes have taken place in local government agencies as well. They convincingly show that policy analysis has played a key role in the rationalisation of governance at this level too. In his contribution, Van der Knaap (Chapter Eight) analyses the role of policy analysis by so-called High Councils of State. Besides the Council of State, a key player in Dutch policymaking and politics, he also looks at the Court of Audit and the National Ombudsman. Part Three includes contributions that focus on the role of more independent institutes for policy analysis, such as committees, public inquiries and consultants. Scholten and Van Nispen (Chapter Nine) describe how the peculiar Dutch system of advisory bodies and think tanks, with the very specific role for ‘planning bureaus’, has come about. Zooming in on four of the most important advisory bodies, they also provide case illustrations how the policy analysis of these institutes played a role in policy developments in different domains. Focusing more on commercial institutes for policy analysis,Van Hoesel (Chapter Ten) shows that the market for policy analysis has shrunk significantly in the last five years, though is still large in comparison to neighbouring countries. Part Four of the book delves into the more political and corporatist dimensions of policy analysis, looking at parties and interest groups in particular. Zaal (Chapter Eleven) provides a unique insight into the growing importance of policy analysis done, or commissioned, by national parliament. She provides a solid overview of how the House of Representatives is reorganising its information position, how it organises parliamentary investigative or inquiry committees, and how this has affected policymaking in a number of cases. Furthermore, providing unique material from their research on political agenda-setting, Timmermans, Van Rooyen and Voerman (Chapter Twelve) provide an original analysis of the role of party-political think tanks in the Dutch context. While the importance of party-political think tanks may be less significant than in countries like the UK and the US, they nonetheless play an 10

Policy analysis in the Netherlands: an introduction

important part in party-political agenda-setting, for instance, by explicating ideological stances. Subsequently, Fenger and Koning (Chapter Thirteen) provide an analysis of a unique phenomenon in the Dutch polity that clearly reflects the consociational history of Dutch politics: Social and Economic Council (Sociaal-Economische Raad; SER). They convincingly describe the role of policy analysis in the production of ‘negotiated knowledge’, which has been particularly influential in the Dutch context. Part Five of the book focuses on the more academic dimension of policy analysis in the Netherlands. First, Thissen (Chapter Fourteen) provides an analysis of policy analysis as instruction in the Netherlands, showing both how strongly policy analysis instruction has been institutionalised at the Dutch universities and how diverse different programmes have become. Next, Bekkers (Chapter Fifteen) provides an overview of the development of academic thinking on policy analysis in the Netherlands. He also positions the development of policy analysis in the context of broader developments in Dutch society, convincingly showing how rational policy analysis is increasingly confronted by and conflicted with incidences of fact-free politics and lay knowledge (a trend observed beyond the Dutch case as well). The last chapter, by Scholten and Van Nispen (Chapter Sixteen), provides a summary and points the way ahead for future research.

Notes In this context, a distinction is also made between a ‘science of policy’ and ‘science for policy’. As with most classifications, this distinction is not watertight.Academics do policyfocused research in the form of contract research, while civil servants conduct research that contributes to the ‘body of knowledge’. 1

The administrative variety of policy analysis institutionalised in the Netherlands in the COBA, which was especially active in the 1970s (Van Nispen and Klaassen, 1995; Mayer, 2007). 2

A policy analyst is a participant instead of an observer, as in the scientific variety (Van Doorn, 1986: 4).

3

An exception to the rule is the Van Beleidsbegroting tot Beleidsverantwoording (VBTB) operation – from policy budgeting to policy accountability – which can be seen as the capstone of the pursuit of efficiency in the public sector.The approach shows similarities with the goal analysis of the COBA and the pursuit of a performance budget, but it would be going too far to talk about ‘old wine in new bottles’ (cf Boorsma et al, 1999). 4

In this case, they need to mobilise such a latent group by means of selective incentives (Olson, 1977: 51). 5

This study is generally regarded as a ‘success story’ since it led not only to decisionmaking, but also to implementation of the selected alternative.

6

11

Policy analysis in the Netherlands

References Bardach, E. (1980) ‘On designing implementable programs’, in G. Majone and E.S. Quade (eds) Pittfalls of analysis, Chichester, New York, NY, Brisbane and Toronto: John Wiley and Sons, pp 138–58. Blume, S.S., Hagendijk, R.P. and Prins,A.A.M. (1991) ‘Political culture and the policy orientation in Dutch social science’, in Wagner (ed) Social sciences and modern states: national experiences and theoretical crossroads. Cambridge: Cambridge University Press, pp 168–90. Boorsma, P.B., Maessen, F.C.M.M. and Schild, J.A. (1999) ‘Van beleidsbegroting tot beleidsverantwoording in relatie tot beleidsanalyse en prestatiebegroting: een historische vergelijking vanuit een beleidsanalytisch perspectief ’, Beleidsanalyse 28(1/2): 15–28. Dror,Y. (1967) ‘Policy analysts: a new professional role in government service’, Public Administration Review XXVII(3): 197–203. Dror,Y. (1984) ‘Policy analysis for advising rulers’, in R. Tomlinson and I. Kiss (eds) Rethinking the process of operational research and systems analysis, Oxford: Pergamom Press: 79–123. Dunleavy, P., Margetts, H., Bastow, S. and Tinkler, J. (2005) ‘New Public Management is dead – long live digital-era governance’, Journal of Public Administration Research and Theory, 16(3): 467–94. Dunn, W.N. (1983) Values, ethics, and the practice of policy analysis, Lexington/Toronto: Lexington Books. Dunn,W.N. (1994) Public policy analysis: an introduction, Englewood Cliffs, NJ: Prentice Hall. Dunn, W.N. (2004) Public policy analysis: an introduction (2nd edn), Englewood Cliffs, NJ: Prentice Hall. Elmore, R.F. (1979) ‘Backward mapping: implementation research and policy decisions’, Political Science Quarterly 94(4): 601–16. Fischer, F. (1993) ‘Policy discourse: the politics of Washington think-tanks’, in F. Fischer and J. Forester (eds) The argumentative turn in policy analysis and planning, Durham, NC: Duke University Press. Fischer, F. and Forester, J. (1993) The argumentative turn in policy analysis and planning, Durham, NC: Duke University Press. Goemans, T. (1988) ‘Beleidsanalyse op weg naar volwassenheid’, in A.F.A. Korsten and Th.A.J. Toonen (eds) Bestuurskunde: hoofdfiguren en kernthemas, Leiden: Stenfert Kroese, pp 343–59. Grindle, M.S. (1980) ‘Policy content and context in implementation’ in M.S. Grindle (ed) Politics and Policy Implementation in the Third World, Princeton, NJ: Princeton University Press, 3–34. Halffman, W. and Hoppe, R. (2005) ‘Science/policy boundaries: a changing division of labour in Dutch expert policy advice’, in S. Maasen and P. Weingart (eds) Democratization of expertise? Exploring novel forms of scientific advice in political decisionmaking, Dordrecht: Kluwer, pp 135–52. Hemerijck,A. (2001) ‘De institutionele beleidsanalyse: naar een intentionele verklaring van beleidsverandering’, in T.Abma and R. in ‘tVeld (eds) Handboek Beleidswetenschap, Meppel: Boom, pp 83–95. Hemerijck, A. (2003) ‘Vier kernvragen van beleid’, Beleid & Maatschappij 30(1): 3–19.

12

Policy analysis in the Netherlands: an introduction

Hoppe, R. (1998) De broosheid van debat en argumentatieve beleidsanalyse, Enschede: Universiteitsdrukkerij. Hoppe, R., Van de Graaf, H. and Van Dijk, A. (1987) ‘Implementation research and policy design’, International Review of Administrative Sciences 53(4): 581–604. Howlett, M. Ramesh, M. and Perl, A. (2009) Studying public policy. Policy cycle and policy subsystems, Oxford: Oxford University Press. Klaassen, H.L. and Van Nispen, F.K.M. (1996) ‘De wildgroei van effectrapportages. Een rapportage’, Bestuurskunde 5(7): 308–16. Knott, J. and Wildavsky, A. (1980) ‘If dissemination is the solution, what is the problem?’, Knowledge: Creation, Diffusion, Utilization 1(4): 537–78. Korsten, A.F.A. (1985) ‘Uitvoeringsgericht ontwerpen van overheidsbeleid’, Bestuur 4(8): 12–19. Kraan D.J. (2010) OECD Value for money study: building on basics. Paper prepared for Working Party of Senior Budget Officials, Paris, 7 July. Lasswell, H.D. (1951) ‘The policy orientation’, in D. Lerner and H.D. Lasswell (eds) The policy sciences: recent developments in scope and method, Stanford, CA: Stanford University Press, 3–15. Leeuw, F.L. (1987) ‘De reconstructie en evaluatie van beleidstheorieën als methode van beleidsonderzoek: achtergronden, werkwijze, toepassing en relevantie’, in M. van de Vall and F.L. Leeuw (eds) Sociaal beleidsonderzoek: differentiatie en ontwikkeling, s-Gravenhage:Vuga. Lijphart, A. (1969) ‘Consociational democracy’, World Politics 21(2): 207–25. Maarse, J.A.M. (1986) ‘De uitvoering van beleid ontworpen’, Beleidsanalyse 15(2): 4–10. Majone, G. (1975) ‘The feasibility of social policies’, Policy Sciences 6(1): 49–69. Majone, G. (1985) ‘From systems analysis to policy analysis’, in H.J. Miser and E.S. Quade (eds) Handbook of systems analysis: overview of uses, procedures, applications and practice, Chichester: Wiley and Sons. March, J. and Olsen, J. (1989) Rediscovering institutions: the organizational basis of politics, New York, NY: Free Press. Mayer, I. (2007) ‘The evolution of policy analysis in the Netherlands’, in F. Fischer, G.J. Miller and M.S. Sidney (eds) Handbook of public policy analysis.Theory, politics, and methods, Boca Raton, London and New York, NY: CRC Press. Meltsner, A.J. (1972) ‘Political feasibility and policy analysis’, Public Administration Review 32(6): 859–67. OECD (2011) ‘Typology and implementation of spending reviews’, OECD Senior Budget Officials Network on Performance and Results, 7th Annual Meeting, Paris, 9-10 November 2011, GOV/PGC/SBO(2011)9. Olson, M. (1977) The logic of collective action. Public goods and the theory of groups, Cambridge, MA, and London: Harvard University Press. O’Toole, L.J. (1986) ‘Policy recommendations for multi-actor implementation: an assessment of the field, Journal of Public Policy 6 (2): 181–210. Patton, C.V. and Sawicki, D.S. (1986) Basic methods of policy analysis and planning, Englewood Cliffs, NJ: Prentice Hall. Patton, M.Q. (2008) Utilization-focused evaluation (4th edn), Los Angeles, CA: Sage. Potman, H.P. (1989) Acceptatie van beleid. Onderzoek naar de Wet geluidhinder ter verkenning van een bestuurskundig begrip, Zeist: Kerckebosch. Radin, B.A. (2000) Beyond Machiavelli. Policy analysis comes of age, Washington, DC: Georgetown University Press. 13

Policy analysis in the Netherlands

Scholten, P. (2011) ‘Constructing Dutch immigrant policy. Research–policy relations and immigrant integration in the Netherlands’, British Journal of Politics and International Relations 13(1): 75–92. Snellen, I.Th.M. (1987) Boeiend en geboeid.Ambivalenties en ambities in de Bestuurskunde, Alphen aan den Rijn: Samsom H.D. Tjeenk Willink. Van de Vall, M. (1987) ‘De waarden-context van sociaal beleidsonderzoek: een theoretisch model’, in M. van de Vall and F.L. Leeuw (eds) Sociaal beleidsonderzoek: een professioneel paradigma, Alphen aan den Rijn: Samsom, pp 35–56. Van Doorn, J.A.A. (1986) ‘Beleidsanalyse en organisatieleer: een poging tot vergelijking en verbinding’, Beleid & Maatschappij XIII(1): 3–19 and 39. Van Nispen, F.K.M. [2014] ‘Policy analysis in times of austerity: the role of spending reviews’, Journal of Comparative Policy Analysis: Research and Practice [forthcoming]. Van Nispen, F.K.M. and Klaassen, H.L. (1995) ‘Drowned as Icarus or risen like phoenix? The flight of policy analysis’, paper delivered to the conference on the Transformation of Values in a Pluriform Society: New Steering Perspectives for Government and Public Policy? Erasmus University of Rotterdam, 21-22 February. Wilson, J.Q. (1980) The politics of regulation, New York, NY: Basic Books.

14

Part One Styles and modes of policy analysis

TWO

Policy analysis in practice: reinterpreting the quest for evidence-based policy Mark van Twist, Rien Rouw and Martijn van der Steen

2.1: Introduction In the past decade, the discipline of policy analysis has regained momentum in the form of attention for ‘evidence-based policy’ (Adams, 2004;Walter et al, 2005; Pawson, 2006; Nutley et al, 2007; Boaz et al, 2008a, 2008b; Head, 2008, 2010; Argyrous, 2009; Banks, 2009; Leigh, 2009; Cartwright and Hardie, 2012). The concept of evidencebased policy was originally developed in the Anglo-Saxon context of British and US public administration (Sherman et al, 1997; Strategic Policy Making Team Cabinet Office, 1999; Solesbury, 2001; De Groot, 2010; Head, 2010). It rapidly expanded to other countries, in part due to the influence of international organisations such as the Organisation for Economic Co-operation and Development (OECD) and the European Union (EU) (Burns and Schuller, 2007; EU, 2007; Head, 2010). In the Netherlands, the rise of evidence-based policy was fuelled by various conclusions of successive parliamentary commissions examining policy failures, which indicated that minimal thorough analysis, and a lack of a sound scientific foundation, were significant, consistent deficiencies that contributed towards policy failure (Bax et al, 2010; De Wree et al, 2010; see also Chapter Twelve on the role of party-political think tanks). The message that resonated in Dutch policy organisations as a result was that successful policy requires sound and rigorous analysis, as well as underlying policy theory (CPB, 2002; Cornet and Webbink, 2004; De Wree et al, 2010). One of the most common interpretations of evidence-based policy at the Dutch national level is that it relies on a quantitative analysis of effectiveness. In other words, it calls for randomised experiments or similar research designs. This often draws comparisons with the double-blind clinical trials that establish the efficacy of new medicines. In both cases, the objective is to establish a causal relationship between an intervention and its intended effects (Cornet and Webbink, 2004). Like proving medicinal effects, the allure of proven policy results is appealing in a political debate. The notion that there exists evidence, proof and, above all, ‘guaranteed value for public money’ resonates strongly in a debate that emphasises the effectiveness of policy. Measurable effect has become an important driver for political debate (Head, 2010). Public funds must be spent efficiently, and in order to do so, policy interventions must be tried and tested. However, effective policy is not the same as policy with a proven effect in a clinical trial experimental setting. Therefore, many scholars and practitioners remain critical about the application of the ‘evidence-based’ standard to policy; they claim that the focus on evidence is embedded too much in principles from different contexts and that the models of clinical trials and large randomised quantitative studies are hardly applicable to real policy for real-world problems (Schon 17

Policy analysis in the Netherlands

and Rein, 1994; Shaw and Shaw, 1997; Colebatch, 2006; Glasby and Beresford, 2006; Biesta, 2007; Bulterman-Bos, 2009).

A long tradition The demand for policy measures to be backed by scientific research is anything but new. Since the Enlightenment, scientific progress has been associated with human and cultural progress, as well as progress in governance and policy (Carson, 2003). Leeuw (2010) describes the important contribution that experimental research made to social progress in the 20th century. Experimental researchers in the fields of sociology and psychology attempted to identify effective interventions, both in terms of the individual and greater society. In doing so, they laid the basis for optimism with regard to government planning, as seen immediately after the Second World War, which led, in turn, to the establishment of various official planning offices. This optimism suffered something of a dent in the 1980s and 1990s, but only temporarily. The late 1990s saw the emergence of a new way of thinking about the relationship between research and policy, which we now term evidence-based policy and utilisation-focused policy analyses. Both approaches place a strong emphasis on knowledge; especially knowledge established in randomised design methodologies and consequently used in practice. For practice, this implies more reliance on evidence; for policy analysis, it implies a more systematic attention to the world of policy practice. The idea behind evidence-based policy is that policies introduced by politicians and decision-makers should rest on scientific evidence, including both positive and negative data, and that policy not supported by proper evidence should be abolished. Connected to this approach is the idea that the relationship between knowledge and policy is linear and instrumental. The basic assumption is that if knowledge is available, it will be known and used by policymakers: good evidence implies good use in policy, with little room for political arguments and gaming over policy decisions.

Contradictory views and practices in evidence-based policy In this chapter, we examine the current state of evidence-based policy in the Netherlands, both in terms of thought and practice. We also review the role played by departments in policy organisations that collect the evidence needed for creating policy.We ask the following research question: how can a productive relationship between research (evidence) and policy be conceptualised, organised and implemented? We deliberately connect conceptualisation with implementation in this chapter. As experience has shown, in order to change the practice of relating evidence to policy, it is first necessary to re-conceptualise the relationship between them. We therefore formulate answers to our research question along two routes. We start with a review of the theoretical debate, both in the Netherlands and worldwide, on evidence-based policy and the role of knowledge in policy. We distinguish between two common interpretations of evidence-based policy: the instrumental and the deliberative. Both interpretations exist in the Dutch debate but can also be found in the international discourse on evidence-based policy (see also Chapter Fifteen). We look at the variant that is most appropriate for strengthening the relationship between research and policy in the current political climate.

18

Policy analysis in practice: reinterpreting the quest for evidence-based policy

After the literature review, we look at the practice of policy research in the Netherlands, to see how the quest for more evidence in policy plays out in practice. In particular, we look at the role of formalised ‘research units’ in local government organisations and examine their attempts to influence policy processes.This is an important choice in our research design, as the ‘locus’ of knowledge, in turn, influences its utilisation. Knowledge is abundant in and around policy organisations, and where knowledge comes from likely carries significant influence as to its use. However, for this study, we look at dedicated research units within policy organisations because they provide us with the clearest perspective on how knowledge eventually contributes to evidencebased policymaking. This part of our study is based on an analysis of the position and role of research departments within local and regional government (Van Daal et al, 2012; see also Chapter Seven on policy analysis by local and regional governments). We consider these units as prime examples exhibiting the broader use of knowledge in policy organisations. The local research units illustrate and reflect changes also taking place at the national policymaking level. Over the past decade in the Netherlands, much more attention has been directed towards channelling results of research into policy processes, partly due to aforementioned parliamentary investigations. For instance, there are policymaking procedures in place that require policymakers to provide explicit documentation of their policies’ basis, with data and evidence. Furthermore, ‘knowledge units’ have been established within ministries charged, among other things, to programme research, to improve relations with research institutions, to enlarge the evidence base and to upgrade the capacity of policymakers to analyse and validate research. The chapter concludes that, reflecting the discussion of evidence-based policy itself, research units and ‘knowledge units’ are expected to apply a far less instrumental approach in supporting policy decisions by adopting a deliberative approach instead – which is very different from their current role. Our findings suggest that the ‘evidence-based’ approach has much to offer to policy, but only if it is done in a less instrumental and more deliberative approach (see also Chapter Five on new modes of policy analysis).That is partly a matter of different methods for collecting evidence, but also suggests a different professional role for researchers and organisational departments that collect evidence, do research and advice policymakers.

Research methods Our assessment of the relationship between research and policy is based on an analysis of data gathered in a large survey study conducted by one of the authors of this chapter (Van Daal et al, 2012). The survey was conducted in 2011/12 among the research units of local and regional authorities, all of which are members of the Netherlands Association for Statistics and Research (VSO), and includes over 90 institutional members. Most respondents were from the larger municipalities in the Netherlands. The directors of all member organisations were invited to take part in the survey, with a response rate of 40%; this provides us with an interesting take on how the community-of-practice of policy researchers in municipalities in the Netherlands perceives recent developments and the foreseeable future of the profession. Although the rate of response was significant, the results do not represent all the municipalities in the Netherlands, and, in fact, over-represent larger municipalities. Nonetheless, 19

Policy analysis in the Netherlands

they provide an important indicator of changing practice in the Netherlands and the embedded or emerging views of the relationship between research and policy in local government. The survey specifically examined the role of the research units and ‘knowledge units’ in the broader policy organisation (our conceptual term for the municipality) that they are part of. We look at three key characteristics: • The positioning of the research units and ‘knowledge units’ within the organisation, that is, their place in the organisational structure and hierarchy, and the influence they are able to exert over the processes and practice of policy formulation and decision-making. This is relevant given the assumption that knowledge and evidence can only be used if policymakers know it. • The professional role and background of the research units’ staff: the manner in which researchers go about their work and their competencies (both in terms of knowledge and expertise, in addition to finer skills such as political sensitivity).This category is relevant for establishing what approach researchers apply when they collect evidence and give advice to policymakers: do they apply an instrumental or a deliberative approach to evidence-based policy? • The performance used to assess whether the research unit and its staff ‘perform well’, for example, the products and recommendations they submit and the degree to which they pursue or promote innovation in their work.This is relevant for exploring what variant of evidence-based policy is most appropriate for strengthening the relationship between research and policy. The results of the survey were discussed during a feedback session attended by respondents, in which conclusions based on preliminary results of the research project were also discussed. In addition to the feedback session, a number of indepth interviews were held with individual researchers from research units and with policymakers working in municipal organisations.

2.2: Theoretical debate on evidence-based policy The concept of evidence-based policy suggests an instrumental relationship between scientific knowledge and policy: evidence from research informs policymakers and helps them make the right decisions (Nutley et al, 2007). In this basic interpretation of evidence-based policy, scientific research offers irrefutable evidence to support the direction of policy. Science produces certain, absolute knowledge that can then be turned into sound and effective policies. According to Hoppe (see Chapter Four), ‘boundary work’ between science and policy can take two forms. In the first view, science itself is dominant and directs policy: technocracy. Politicians act on the results of scientific research without further consideration: they do as the evidence tells them to. In the second view, policy is dominant over research, and policymakers themselves decide when to ask for the advice and help of scientific research. Hoppe refers to this as the engineering model. Politicians and policymakers retain the services of scientists in order to design the intended policy based on scientific knowledge. In both views, the relationship is not only instrumental, but also linear: knowledge and policy are produced in relatively isolated worlds, and there is a hierarchy, with ‘pure’ knowledge at the top and the practical application of knowledge far lower on the scale (Hoppe, 2002; Regeer and Bunders, 2006). 20

Policy analysis in practice: reinterpreting the quest for evidence-based policy

The proponents of evidence-based policy generally work under the assumption that there is another hierarchy, that of scientific methodologies (Cornet and Webbink, 2004; Rawlins, 2008). Within the established tradition of evidence-based policy, there is a strong emphasis on review studies and evaluations of effectiveness, which are intended to show a causal relationship between (policy) interventions and effects. Most of these studies take a randomised design approach, or randomised controlled trial (RCT), in which an intervention group and a control group are selected at random.This form of experiment is seen as the gold standard, producing the highest and most reliable form of knowledge, improved only by subsequent meta-analyses of several RCTs. Qualitative research and the resultant knowledge are regarded as somewhat inferior in terms of analysis strength. Good evidence is considered the result of repeated RCTs, which distinguish it from other, less rigorous methods.This is, for instance, the starting point for analyses of the influential Netherlands Bureau for Economic Policy Analysis (Centraal Planbureau; CPB). The CPB validates a range of policy interventions as ‘promising’ and as ‘contributing to economic growth’ (CPB, 2012). As a result of the CPB’s work, the range of policy proposals in political programmes has been confined to what it has proven effective according to its standards, and, as a consequence, certain domains have experienced the execution of a steadily increasing number of experiments. However, in spite of the academic quality of evidence’s hierarchical system of judgement, Nutley et al (2007: 15) point at the gap between the rhetoric of evidencebased policy and local practice on the ground. Practice does not reflect the explicit or implicit doctrines that are the essence of the evidence-based policy approach; there is much talk about evidence-based policy, and there are many sincere attempts to introduce the concept into policymaking, but what actually takes place in practice hardly resembles the RCT model. This is, at least in part, because the complexity of real-world policy problems is such that it can never be ‘pinpointed’ in randomised trials. In the Netherlands, this argument is used by many pedagogical scientists who stress the unpredictable reality of classrooms. Another stream of critique (see Van Montfort et al, 2008a) focuses more on the practical world of policymaking itself, where matters of politics, path dependency and institutionalisation mitigate the role of scientific evidence to merely one of the many (and minor) factors involved in decision-making. Evidence can support policy, but evidence does not control policy. In the remainder of this chapter, we will elaborate on both points. First, we examine how to best quantify the level of ‘science’ in evidence that forms the basis of policyoriented research. Second, we delve into other, less instrumental models for the relationship between evidence and policy. At first glance, such models relativise the role of evidence in policy. Upon further examination, however, they also create new perspectives on the use of evidence in the context of policy.

Mixed methods: there is no gold standard for evidence-based policy The first question that requires an answer is whether scientific research can actually produce ‘certainty’ in the complex social context of policy. Essentially, research is a deliberative activity that, all things considered, is principally provisional: science constantly strives to refute and disprove earlier theories (Cook, 2003). Does this also hold true for an applied science such as policy analysis? Here, research can only establish what has worked, not what is working now or will do so in the future (cf Biesta, 2007). 21

Policy analysis in the Netherlands

A second critique of the ‘pure’ model of evidence-based policy is that the instrumental view of evidence-based policy postulates a neutral relationship between objectives and resources, while in the real world, interventions are always valueladen (Webb, 2001; Biesta, 2007). This leads to a reduction in ‘varied reality’. In fact, research itself is never entirely neutral. Quite the contrary, it is essentially performative. Research not only describes and reflects reality, but constructs and changes it. In that sense, research is outdated as soon as it is added to a system, since its introduction alters the system itself. Third, randomised research is not always capable of meeting expectations for reasons of methodology. To increase internal validity, an important concession to practical applicability needs to be made. The effect of interventions in social practice largely depends upon context. No two situations are identical; thereby, links and connections established at the meta-scale are of little value (see Shaw and Shaw, 1997;Webb, 2001). It is interesting to note that in medical research, often cited as the role model for evidence-based policy, there is an ongoing debate over the degree to which the results of randomised studies can be extrapolated and subsequently introduced on a greater scale. Rawlins (2008) suggests that the subjects in randomised trials are rarely, if ever, truly representative of the general population, and so it is far from certain whether the results of the trial can then be applied to all patients with a certain condition. Therefore, researchers themselves increasingly agree that, as tempting as the idea may be, there is no actual gold standard for policy-relevant evidence, not because it has not yet been found, but because it does not exist. Quasi-experimental design remains the most suitable means of establishing causal relationships, but it has important limitations that cannot be overcome by simply altering experiments’ framework (Rawlins, 2008). As Leeuw and Vaessen (2009: x) assert in their study of impact evaluation in development cooperation: ‘No single method is best for addressing the variety of questions and aspects that might be part of impact evaluations’, going on to state that, ‘in doing impact evaluations there is no “gold standard”’ (Leeuw and Vaessen, 2009: xiii). Rawlins (2008: 2159) makes the same point even more forcefully: ‘Hierarchies attempt to replace judgment with an over-simplistic, pseudo-quantitative assessment of the quality of the available evidence’. Rawlins argues that policymakers always need to assess whether evidence presented to them is ‘fit for purpose’. Scientific rigour is not enough to produce practically relevant and applicable knowledge. Taking these considerations into account, the Dutch Ministry of Education, Culture and Science, for instance, has formulated a combined research design in its framework for policy research that guides the assessment of research tenders.According to this framework, a sound, effective research design consists of a combination of quantitative research on causal relations with qualitative research into explanatory mechanisms. Another example from the Netherlands is the work of the Research and Documentation Centre (Wetenschappelijk Onderzoek en Documentatiecentrum; WODC) of the Ministry of Security and Justice. One of the characteristics of its research programme is an emphasis on the reconstruction of ‘policy theories’ or theories of change as a crucial component of research on the effects of interventions (Klein Haarhuis and Hagen, 2009).

22

Policy analysis in practice: reinterpreting the quest for evidence-based policy

The relationship between evidence and policy: ‘power versus puzzle-solving’ A second problem that plagues interpretation of evidence-based policy is that the model hardly does justice to the complex practice of policymaking.This is not so much an issue regarding the method of research, but more a result of assumed instrumental relationships between knowledge and policy – either in the technocracy or engineering models (Hoppe, 2002). It is possible that there was once a time when public administration theory could safely make such assumptions. The instrumental approach fits into the model of a linear policy cycle in which policy development is undertaken in a number of clearly defined phases, from preparation to evaluation and subsequent adjustment. Numerous analyses have since shown that, in practice, the process of political decision-making can be extremely chaotic; it is far from linear or rational (Van de Donk and Hemerijck, 2007; Michels, 2008). Of course, policymakers themselves are all too aware of this, but they continue supporting the rational decision-making model as ideal (Rouw, forthcoming). More importantly, knowledge is not incorporated into decision-making processes in a value-free form, but becomes another weapon in the battle between political parties (Van Montfort et al, 2008b). Knowledge is used strategically, as part of a story, argument or ‘frame’. A politician uses this frame to dominate and control public discourse to their advantage. Examples of frames include ‘street terrorism’, ‘the human dimension’ and ‘bureaucracy versus professionals’. They are expressed as statements and political messages in which certain facts and information are pushed to the fore, while other facts and information are disregarded in a practice known as ‘spinning’. Alongside these frames,‘beliefs’ also influence the use of scientific analysis, in particular, the beliefs held by the majority of the electorate, in which the future of any given politician lies. The value of knowledge as a political weapon explains why knowledge is used selectively, and, at times, is not part of the process at all, as the recently oft-heard term ‘fact-free politics’ suggests (see Van Montfort et al, 2008a). In many cases, the actual decision is made before supporting evidence is available (see Knott and Wildavsky, 1980). Some solutions are looking for problems, and evidence is sometimes merely needed to justify and legitimise a political or ideological – or highly pragmatic – decision about policy. In most cases, knowledge and research are part of the package of a policy proposal, but in roles that vary widely: at times, policy follows knowledge; at other times, knowledge gathers around a ready-made policy. The political nature of knowledge use in policy is at least partially influenced by the nature of any given policy issue. Not all issues provide the same space for political manoeuvring with evidence. Not all issues are equally politicised. Much depends upon whether there is agreement regarding values at stake and whether there is sufficient knowledge available to address the issues (see, eg, WRR, 2006). If knowledge is reasonably conclusive and there is consensus about the political values involved, research can be incorporated into policy processes in a highly instrumental form. We then see a ‘tamed problem’. A problem is ‘untamed’ if knowledge is not conclusive and there is significant disagreement about political values.The US researcher Roger A. Pielke Jr (2007: 41) summarises the distinction between tamed and untamed problems by differentiating between ‘Tornado politics’ and ‘Abortion politics’. If government wants to learn how to improve its response to tornado threats, it is useful to have knowledge about the 23

Policy analysis in the Netherlands

storm’s strength, position, trajectory and pace. Hard knowledge is needed to support decisions, especially in critical instances, for example, when deciding whether to order a full-scale evacuation or not. Politicians and policymakers will be inclined to look to science to provide them with strategies for dealing with the problem of the upcoming tornado – and set aside their political beliefs in order to do so. Contrary to the tornado, abortion politics relies far less on science and is more infused with values and beliefs. Factual information does not lead the debate, which is instead characterised as a more fundamental conflict pitting ideological views on abortion against each other. The issue is essentially political, and facts will be used as mere ‘weapons’ in the political battle over policy. The instrumental and linear approach to the relationship between evidence and policy hardly applies to untamed, ‘wicked’ problems. In order to apply knowledge to such highly politicised issues, a joint effort to identify and define the problem, and then devise a solution, is necessary. Power alone cannot solve wicked issues, since ‘good solutions’ require at least partial grounding in sound theory. Research can provide answers, but it requires careful and deliberative processes in order to be accepted.Van de Donk and Hemerijck (2007: 334) stress that wicked problems require as much ‘power play’ as they do ‘puzzle-solving’: ‘Politics finds its sources not only in power but also in uncertainty – people collectively wondering what to do’. The puzzlesolving approach allows for the possibility that intervention will have unforeseen consequences, which is generally the case in any complex, real-world situation (Van Twist and Verheul, 2010).

2.3: An empirical perspective: the practical application of knowledge in policy? With the theoretical review in mind, it is interesting to take a closer look at the practice of knowledge production in policy organisations and at the attempts and processes in which knowledge workers in policy organisations bring their products to policy workers (see Colebatch et al, 2010; Van der Steen and Van Twist, 2012). First of all, it is interesting to look at the positioning of knowledge units within the formal and relational network of the organisation. Where is the unit located and at what level are its most frequent and prominent interactions? Empirical analysis of the situation in 36 of the larger municipalities in the Netherlands shows that research departments are often centrally positioned within their respective organisations (Van Daal et al, 2012). However, while strategically positioned at the heart of the organisation, at its strategic core, most of their contacts work at the tactical and operational level of the organisation. Apparently, most of the interaction between knowledge and policy takes place at the operational level in the organisation. Strategic contacts do take place but only occasionally and on an ad hoc basis.What stands out is that knowledge is used in the policymaking process, but that there is little direct access of knowledge workers to the arena of decision-making itself.There is an indirect presence of knowledge in policy: others than the researchers themselves present the findings and introduce evidence into the policy process. Knowledge and policy are produced in relatively isolated worlds. This has important consequences for both worlds: knowledge units can hardly develop an understanding of what policymakers

24

Policy analysis in practice: reinterpreting the quest for evidence-based policy

want and need, whereas policymakers have little understanding of what knowledge workers do, how they work and what sort of questions they can ask knowledge units. If we look at the work staff members of the research units in the 36 municipalities, we notice significant attempts at change in recent years. Advice and policy evaluation receive more attention, at the expense of the more traditional tasks of gathering statistical information and data management. In general, we can state that staff members in research units engage more in activities aimed at bringing knowledge to the policy process than they do merely developing new knowledge and waiting for its selection and application to policy. In that sense, researchers take a more proactive role in the relationship between knowledge and policy, at the expense of the time they invest in actually gathering data and doing statistical analysis.Their organisational position has not changed, but the work has. They spent less time doing analysis and invest more time in interaction with their relevant policy environment. Obviously, their primary occupation remains focused on ‘undertaking research’, but the focus on interaction with policymakers and the application of insights has certainly increased. This renewed focus on strategic advice and the proactive role of researchers in their relationship with policymakers is further underscored if managers and staff members of research units are asked about what they think will be the most likely changes to their profession in upcoming years (Van Daal et al, 2012). They feel that direct relevance to policy will become a more important aspect of their work. They strive to become more relevant to policymakers in policy processes and foresee a move towards a more intense interaction with the strategic core of the organisation. In order to achieve this, managers and staff members of research units expect it to be necessary to focus less on academic rigour and more on policy relevance. Focus and attention will go more towards picking the relevant topics than defining the best possible methods for empirical (statistical) research. Respondents also expect a shift from mainly quantitative methods towards an increasingly mixed branch of qualitative and quantitative research influenced by the issues at hand, and the manner in which research will be applied towards those issues. In addition, staff members expect their work to become increasingly interaction-oriented, with an expanded role for interaction with policymakers and other ‘users’ of their products. Furthermore, the current output-oriented approach that prioritises producing reports over examining the possible use of knowledge will be replaced by an approach in which policymakers co-produce or at least participate in the research. The managers and staff members of the research units in 36 of the larger municipalities in the Netherlands expect that the nature of their work will change. They expect a more interactive process of research, where external parties will conduct selected research and outsiders will be temporarily invited based on expertise to strengthen project teams. Research will remain relatively autonomous, but there will be more room for the joint assessment of findings. In that sense, they expect the research process and the units to ‘open up’. The common thread connecting these expected changes in the nature of their work is the importance of the connective, interactive and deliberative capacity and competencies of research staff.They expect their work to become more relational and interactive, and, in that sense, less about the research itself.They argue that in order for the research function to fulfil its potential, it must seek closer contact with internal and external stakeholders, and take a proactive role in policy processes. That implies less emphasis on the independent and autonomous positioning of knowledge workers, 25

Policy analysis in the Netherlands

put deliberately at a distance from where the policymaking happens. Knowledge workers pro-actively seek interaction with policymaking, allow policy makers into the process of research, and at the same time themselves participate actively in the fact-finding phase of policymaking. Both are currently considered intrusive and harmful, but – as our respondents expect – they will become more important and ‘normal’ in the coming years. Consistent with anticipated trends concerning the role of internal research units, the managers and staff members of the research units we interviewed about the relation between research and policy considered connective competencies and communication skills as likely to become more important in their work. Networking abilities are also part of the new profile for researchers, just like external awareness, presentation skills and problem analysis. Finally, we looked at the performance demanded of the research units and ‘knowledge units’ in 36 of the larger municipalities of the Netherlands. The managers and staff members we interviewed considered innovation in research methods as absolutely essential for the future of the profession. That implies diminished dominance by the academic method, but without loss of rigour. Research has to be both academically sound and practically relevant, which, in turn, requires the development of methods and techniques that meet these goals while maintaining the standards of both policy and research. Also, the knowledge workers in the research units foresee the need to invest in research methods that allow the units to come up with proper answers to sudden, unforeseen quandaries.As the research units move more towards the strategic policy core of the organisation, they also become subject to a more dynamic agenda and the sometimes ad hoc emergence of pressing questions. Again, the goal is to determine and develop methods that can swiftly produce answers, but that also uphold certain academic and scientific standards of quality in a context of evidence-based policymaking. It is again necessary that they operate successfully in both worlds: in the dynamic context of politics and policy; and in the traditional and institutionalised context of academic research. The empirical results from our survey of the perspectives of municipal researchers further reflect general developments in Dutch policymaking in relation to policy research. Much more attention is given to intensive communication and close connections between researchers and policymakers throughout the research and policymaking process. Furthermore, as we have already observed, in certain fields like education policy and law enforcement, the combination of rigorous quantitative research with qualitative research is emphasised.

2.4: Conclusion and discussion: towards deliberative evidencebased policy The empirical findings from our survey underline the debate about evidence-based policy: what originally began as a model rooted in clinical trials is now moving towards a more deliberative and less instrumental approach.We have seen that in the Netherlands, employees in research units of municipalities are moving away from the classic role and responsibilities of a researcher, and towards a more connective and interactive role, as an ‘adviser’ or ‘strategist’.The basic type of advice provided remains similar in orientation, however, and is uniformly based in knowledge and research. 26

Policy analysis in practice: reinterpreting the quest for evidence-based policy

Rigour and sound methodology remain important, but the Dutch practitioners in our survey argue that they see that evidence does not automatically find its way into policy anymore. They see an important role for research and evidence in policy, but also stress the need to actively connect their findings to the arenas of decision-making. That may require a change in the attitudes of policymakers, but it is very interesting that the researchers in our data set foresee, first and foremost, a different role for themselves. The community of practice of policy researchers in the Netherlands seems to be in transition. For the near future, the practitioners expect more attention to deliberative and interactive activities, and perceive that as an important change in their work. However, this is not a simple matter of the future or an expected refocusing of research attention in policy organisations; practitioners also say that the profession has already moved more towards such activities and competencies over the last couple of years. In that sense, they perceive an acceleration of developments, but not a dramatic shift in the work and priorities of researchers in policy organisations, nor in evidence’s role in policymaking. The shift in researchers’ attention is indicative of new developments in evidencebased policy. Evidence-based policy is still about increasing the role of knowledge, research and evidence in policy processes, but, in practice, it has moved away from the classical linear models towards a more interactive model. The most important challenge for the knowledge workers in the municipalities that we researched will be to find methods and ways of working that fit the needs of both the world of policy and the basic properties of sound research. Policy is helped not by ‘bad science’, but by good researchers that can both produce good-quality research and bridge the gap with strategic decision-makers in policy processes. That trend has important consequences for policy research, as a more deliberative approach to evidence-based policy, in part, implies that researchers should interact more with policymakers, but it also means that the research itself becomes more deliberative. In the remainder of this section we will look into what that deliberative turn in evidence-based policy means for the practice of policy research, and for researchers. First, a deliberative approach to evidence-based policy requires a combination of research methods. A useful policy analysis requires insight into the cause and effect of suggested policies, but it also requires information about the context of a policy intervention. Causality in policy is highly contextual and, as a consequence, often very local. It remains important to ask ‘What works?’, but the answer for most issues cannot expand beyond aspects that are local and unique to a particular circumstance – instead of being generally applicable to every situation.This is the ‘realistic evaluation’ model developed by the British sociologists Pawson and Tilly; the model includes the measurement of outcomes and a thick description of both explanatory mechanisms and the context of an intervention. This type of design demands a combination of theoretical, quantitative and qualitative research. The practice of evidence-based policy in the Netherlands seems to be moving towards the realistic evaluation model.A good example of the necessity for qualitative insights is a study of policy for ‘very weak schools’ in the Netherlands (Van Twist et al, 2013). At the macro-level, policy for very weak schools is highly effective. The number of schools classed as ‘weak’ or ‘very weak’ has fallen in recent years. However, at the micro-level, there is very little insight into why and how ‘very weak’ schools improve. In fact, the quantitative study of the micro-practice of policy interventions in very weak schools could not find causal mechanisms to explain outcomes. Only 27

Policy analysis in the Netherlands

after a very close and qualitative analysis of what happened in the rich and local context of individual schools were local explanations for effects discovered. The qualitatively measurable ‘success’ of policy could only be explained by very local and qualitative analysis. Mixed measures are necessary for a proper understanding of if and how policy works. Furthermore, such a rich level of understanding is necessary for application in policymaking, even though it comes at the relative price of a loss of generalisable findings. Second, the deliberative interpretation of evidence-based policy means that research cannot produce ready-made action plans or recipes for success, but it can provide insights that may improve the decision-making process. In essence, they do not provide the answers for policymakers, but they can add information to or reflect on the debates of policymakers. The findings give an interesting insight into where the ‘Dutch case’ is headed. In the Netherlands, this development seems to be in a deliberate direction: research as one ingredient in a rich, open and purposeful process of policymaking, alongside other ingredients that are, in some cases, highly irrational, unscientific and contrary to the principles of academic research. That is often seen as a flaw of policymaking, but it is better perceived as an evident characteristic of the nature of policymaking. Policy is never founded on evidence from research only. Good judgement requires more than knowledge. Sackett et al (1996: 71) state that evidence-based medicine is ‘the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients.The practice of evidence-based medicine means integrating individual clinical expertise with the best external clinical evidence from systematic research’. Davies (quoted in Sutcliffe and Court, 2005: 1) describes evidence-based policy as ‘the integration of experience, judgement and expertise with the best available evidence of systematic research’. Good policy does not come from good evidence, but from high-quality processes in which the best available evidence is interpreted as well as possible by professional policymakers. Evidence is one element of a much broader and more heterogeneous equation. The typically Dutch phenomenon of so-called knowledge chambers is a way of organising such processes. Various ministries have established such a knowledge chamber as a means of interaction between the top officials of the ministry and eminent researchers on state-of-the-art research about policy issues (Rouw, 2011). Third, the interactive and deliberative role of research – and researchers – in policymaking requires the acceptance of politics as an asset rather than a distraction in policy processes. The role of evidence in policymaking cannot be to impose decisions upon politicians, or otherwise take over the decision-making and political choice of politicians. Evidence feeds into a process of decision-making, but is then subject to what is essentially political. Knowledge expands the ‘horizon of possibilities’ and the researcher can act as an ‘honest broker’ (Pielke, 2007), but the decision is ultimately left to politicians. In most cases, research will reveal a range of possible interventions, each with advantages and disadvantages, desirable and adverse effects, and different costs and returns.A proper assessment, with good judgement, is required to determine which of these interventions offers the best balance between pros and cons, and costs and returns. There must also be a careful and reasoned assessment of whether an intervention that has been effective in one context can be expected to have the same effect in another. In this way, evidence can add significant quality to policymaking. On a small scale, the spirit of experimentalism is currently spreading among policymakers in various Dutch ministries: several interventions are not only 28

Policy analysis in practice: reinterpreting the quest for evidence-based policy

being piloted, but also evaluated as rigorously as possible before their implementation across the country. In this respect the Netherlands is a front-runner, at least when compared to many other European countries.

Towards a more deliberative relationship between knowledge and policy How can we re-conceptualise the relationship between evidence and policy? What emerges when evidence and policy grow closer to one another and become more deliberative and interactive processes, while, at the same time, they have to remain true to their particular founding principles and basic rules? What form will the ‘new in-between’ of science and policy take? What are its design requirements? Our findings suggest several principles for reorganising the relationship between research and policy. First of all, interaction between the worlds of knowledge and policy should develop from what is, at present, a primarily project-based approach towards an ‘ongoing dialogue’ (Habermas, 1984; CHSRF, 2004) or ‘sustained interactivity’ (Sutcliffe and Court, 2005). Policymakers and researchers meet only when there is a specific result, intermediate product or end result to be discussed, but there is much to gain from more consistent interaction. Moreover, researchers and policymakers should be linked directly to each other instead of research being interpreted and edited by intermediary institutions or officials. Although brokerage between research and policy or practice is necessary in many cases, it requires a flexible and light organisational mode. At the national level, various ministries have established knowledge directorates to play this role. Second, as Nutley et al (2003) argue, the dialogue concerning research and policy should become more polygamous. In order to maintain a rich discourse, it is important to invite more perspectives to the table. Research should not be commissioned to just one selected partner, but should instead include several partners simultaneously as part of the conversation with policy. Third, an ongoing, rich and heterogeneous conversation between research and policy requires a certain capacity on the side of policymakers to understand the research process and its results, not in order to become researchers themselves, but so that they can interpret results correctly and apply them in their own practice. Policymakers must become the ‘intelligent customers’ of research (Campbell et al, 2007; cf Bowen and Zwi, 2005). This calls for further training and education, not in the profession of becoming a good researcher, but in the skills and attitudes necessary to deal with scientific evidence in complex policy processes. Several universities and training institutes have already developed such training programmes. All in all, putting policy analysis into practice not only requires evidence based on various sources of research, both quantitative and qualitative, but also demands flexible and dynamic relationships between the different communities (Caplan, 1979; Dunn, 1980) of researchers and policymakers, based on mutual respect and the genuine desire to understand each other’s worlds and views.That requires learning by policymakers and researchers, not to learn more about the world as an empiric object of a scientific study or policy, but to develop a deeper understanding of what moves their counterpart in a mutual relationship. In that sense, strengthening the evidence base of policy requires as much introspection and interaction as it does additional research and theory development.

29

Policy analysis in the Netherlands

References Adams, D. (2004) ‘Usable knowledge in public policy’, Australian Journal of Public Administration 63(1): 29–42. Argyrous, G. (ed) (2009) Evidence for policy and decision-making, Sydney: UNSW Press. Banks, G. (2009) ‘Evidence-based policy-making: what is it? How do we get it?’, ANZSOG Public Lecture, 4 February.Available at: http://www.pc.gov.au/speeches/ cs20090204 Bax, C., De Jong, M. and Koppejan, J. (2010) ‘Implementing evidence-based policy in a network-setting: Dutch road safety policy in a shift from a home to an away match’, Public Administration 88(3): 871–84. Biesta, G. (2007) ‘Why “what works” won’t work: evidence-based practice and the democratic deficit in educational research’, Educational Theory 57(1): 1–22. Boaz, A., Fitzpatrick, S. and Shaw, B. (2008a) Assessing the impact of research on policy: a review of the literature, London: King’s College and Policy Studies Institute. Boaz,A., Grayson, L., Levitt, R. and Solebury,W. (2008b) ‘Does evidence-based policy work? Learning from the UK experience’, Evidence & Policy 4(2): 233–53. Bowen, S. and Zwi, A.B. (2005) ‘Pathways to “evidence-informed” policy and practice: a framework for action’, PLoS Medicine, 2(7): 600–5. Bulterman-Bos, J. (2009) ‘Naar een nieuwe opzet van de onderwijswetenschappen?’, in R. Rouw, D. Satijn and T. Schokker (eds) Bewezen beleid in het onderwijs, Den Haag: Ministerie van OCW. Burns, T. and Schuller, T. (eds) (2007) Evidence in education: linking research and policy, Paris: Centre for Educational Research and Innovation (CERI), OECD. Campbell, S., Benita, S., Coates, E., Davies, P. and Penn, G. (2007) Analysis for policy: evidence-based policy in practice, London: Government Social Research Unit. Caplan, N. (1979) ‘The two communities’ theory and knowledge utilization’, American Behavioral Scientist 22: 459–70. Carson, W.G. (2003) ‘Evidence based policy and crime prevention in Victoria’, unpublished paper. Cartwright, N. and Hardie, J. (2012) Evidence-based policy: a practical guide to doing it better, Oxford, CT: Oxford University Press. CHSRF (Canadian Health Services Research Foundation) (2004) What counts? Interpreting evidence-based decision-making for management and policy, Ontario: CHSRF. Colebatch, H.K. (ed) (2006) Beyond the policy cycle, Sydney: Allen and Unwin. Colebatch, H., Hoppe, R. and Noordegraaf, M. (2010) Working for policy, Amsterdam: Amsterdam University Press. Cook, T.D. (2003) ‘Why have educational educators chosen not to do randomized experiments?’, Annals of the American Academy of Political and Social Science 589: 114–49. Cornet, M. and Webbink, D. (2004) ‘Lerend beleid: het versterken van beleid door experimenteren en evalueren’. Available at: www.cpb.nl CPB (Netherlands Bureau for Economic Policy Analysis) (2002) De pijlers onder de kenniseconomie: opties voor institutionele vernieuwing, Den Haag: Centraal Planbureau. CPB (2012) Charted choices 2013–2017, Den Haag: Centraal Planbureau. De Groot, H. (2010) ‘Evidence-based public management’, speech, 3 June, Enschede, University of Twente. De Wree, E., Devroe, E., Broer, W. and Van der Laan, P. (eds) (2010) Evidence based policing, Apeldoorn: Maklu-Uitgevers.

30

Policy analysis in practice: reinterpreting the quest for evidence-based policy

Dunn,W.N. (1980) ‘The two communities metaphor and models of knowledge use’, Knowledge: Creation, Diffusion, Utilization 1: 515–36. EU (European Union) (2007) ‘Towards more knowledge-based policy and practice in education and training’, Commission Staff Working Document.Available at: http:// bookshop.europa.eu/en/towards-more-knowledge-based-policy-and-practice-ineducation-and-training-pbNC7807296/ Glasby, J. and Beresford, P. (2006) Who knows best? Evidence-based practice and the service user contribution’, Critical Social Policy 26(1): 268–84. Habermas, J. (1984) Theory of communicative action (trans Thomas McCarthy), Boston, MA: Beacon Press. Head, B.W. (2008) ‘Three lenses of evidence based policy’, Australian Journal of Public Administration 67(1): 1–11. Head, B.W. (2010) ‘Reconsidering evidence-based policy: key issues and challenges’, Policy and Society 29(2): 77–94. Hoppe, R. (2002) Van flipperkast naar grensverkeer.Veranderende visies op de relatie tussen wetenschap en beleid, Den Haag: AWT. Klein Haarhuis, C.M. and Hagen, L.L.C (2009) Toetsen en verbinden, Den Haag:WODC. Knott, J. and Wildavsky,A. (1980) ‘If dissemination is the solution, what is the problem?’ Science Communication, June 1980, 537–78, doi:10.1177/107554708000100404. Leeuw, F. (2010) ‘On the contemporary history of experimental evaluations and its relevance for policy making’, in O. Rieper, F.L. Leeuw and T. Ling (eds) The evidence book, New Brunswick, NJ, and London: Transaction Publishers. Leeuw, F. and Vaessen, J. (2009) Impact evaluation and development. NONIE guidance on impact evaluation, Washington, DC: Network of Networks on Impact Evaluation. Leigh, A. (2009) ‘What evidence should social policymakers use?’, Economic Roundup 1: 27–43. Michels, A. (2008) ‘Kennis en conflict in beleidsprocessen’, Bestuurskunde 2: 5–14. Nutley, S., Davies, H. and Walter, I. (2003) ‘Evidence-based policy and practice: cross-sector lessons from the United Kingdom’, Social Policy Journal of New Zealand, Online 1 June 2003. Nutley, S.,Walter, I. and Davies, H. (2007) Using evidence: how research can inform public services, Bristol: Policy Press. Pawson, R. (2006) Evidence-based policy: a realist perspective, London: Sage. Pielke Jr, R.A. (2007) The honest broker. Making sense of science in policy and politics, Cambridge: Cambridge University Press. Rawlins, M. (2008) ‘De testimonio: on the evidence for decisions about the use of therapeutic interventions’, The Lancet 372(20/27): 2152–61. Regeer, B.J. and Bunders, J.F.G. (2006) Kenniscocreatie: samenspel tussen wetenschap en praktijk, Den Haag: RMNO. Rouw, R. (2011) Gevoel voor bewijs. Naar vloeiende verbindingen tussen kennis en beleid, Den Haag: NSOB. Rouw, R. (forthcoming) ‘Een beleidspraktijkperspectief op het gebruik van evaluaties’, Achtergrondstudie voor de WRR. Sackett, D.L., Rosenberg, W.M.C., Muir Gray, J.A., Haynes, R.B. and Richardson, W.S. (1996) ‘Evidence based medicine: what it is and what it isn’t’, British Medical Journal 312: 71–2. Schon, D.A. and Rein, M. (1994) Frame reflection: toward the resolution of intractable policy controversies, New York, NY: Basic Books. 31

Policy analysis in the Netherlands

Shaw, I. and Shaw, A. (1997) ‘Keeping social work honest: evaluating as profession and practice’, British Journal of Social Work 27: 847–69. Sherman, L.W., Gottfredson, D., MacKenzie, D., Eck, J., Reuter, P. and Bushway, S. (1997) ‘Preventing crime: what works, what doesn’t, what’s promising. A report to the United States Congress’. Available at: https://www.ncjrs.gov/works/ Solesbury, W. (2001) ‘Evidence based policy: whence it came and where it’s going’, ESRC UK Centre for Evidence Based Policy and Practice Working Paper 1. Strategic Policy Making Team Cabinet Office (1999) Professional policy making for the twenty first century, London: Cabinet Office. Sutcliffe, S. and Court, J. (2005) Evidence-based policymaking: what is it? How does it work? What relevance for developing countries?, Overseas Development Institute. Turnhout, E., Hisschemöller, M. and Eijsackers, H. (2008) ‘Science in Wadden Sea policy: from accommodation to advocacy’, Environmental Science & Policy 11(3): 227–39. Van Daal, A., Misdorp, P. and Van Twist, M. (2012) Van waarheidsvinding naar publieke waarde, Den Haag: NSOB. Van de Donk, W. and Hemerijck, A. (2007) ‘Leren voor beleid: over regeren en vooruitdenken’, P. den Hoed and A.-G. Keizer (eds) Op steenworp afstand, Den Haag: WRR. Van der Steen, M. and Van Twist, M. (2012) ‘Beyond use: evaluating foresight that fits’, Futures, 44: 475–86. Van Montfort, C., Michiels, A. and Van der Steen, M. (2008a) ‘Een klassiek probleem heroverwogen’, Bestuurskunde 2: 2–4. Van Montfort, C., Michiels,A., andVan der Steen, M. (2008b) ‘Tussen willen en weten: het cynisme voorbij’, Bestuurskunde 2: 56–60. Van Twist, M. and Verheul, W.J. (2010) ‘Onvoorziene opbrengsten’, B&M 37(4): 308–18. Van Twist, M.,Van der Steen, M., Kleiboer, M., Scherpenisse, J. and Theisens, H. (2013) Coping with very weak primary schools: toward smart interventions in Dutch education policy, Paris: OECD. Walter, I., Nutley, S. and Davies, H. (2005) ‘What works to promote evidence-based practice? A cross-sector review’, Evidence and Policy 1(3): 335–64. Webb, S.A. (2001) ‘Some considerations on the validity of evidence-based practice in social work’, British Journal of Social Work 31: 57–79. WRR (Wetenschappelijke Raad voor het Regeringsbeleid) (2006) Lerend beleid, Amsterdam: AUP.

32

THREE

Policy analysis in networks: the battle of analysis and the potentials of joint fact-finding Arwin van Buuren and Joop Koppenjan

3.1: Introduction The Netherlands is renowned for its policymaking style of accommodation and consensus seeking. This policy style is rooted in a long tradition of polder politics in a highly fragmented, decentralised system, collaboration among the elites of various denominations, and neo-corporatist negotiations among the state, employers and unions (see also Chapter Eight). The term ‘polder politics’ refers to the emergence of democratic decision-making among inhabitant of polders, interdependent in their need to build and maintain dikes for their polders in order to stand fast in their battle against a common enemy: water (Lijphart, 1968; Hendriks and Toonen, 2001). However, this policy style of negotiation and consensus building has not been characteristic for all Dutch policy areas. Especially in policy areas in which specialised knowledge was required to solve problems, and in which stakeholders were absent or did not have an institutionalised position, different ways of policymaking evolved (Van Putten, 1982). Alongside the reputation of polder politics, the Netherlands also has a strong tradition of policymaking based on the knowledge of public scientific institutions (Mayer, 2007). Especially in the field of water management, spatial planning, health, the environment and infrastructure, well-known knowledge institutions like Deltares, Netherlands Organisation for Applied Scientific Research (Nederlandse Organisatie voor Toegepast Natuurwetenschappelijk Onderzoek;TNO), Netherlands Bureau for Economic Policy Analysis (Centraal Planbureau; CPB) and Alterra have played and still play an authoritative role. Many policies in these areas are actually based on the expertise and the analyses that these institutions present to policymakers. The influential analysis of the RAND Corporation that underlies the decision for the Easter Scheldt Barrier is a powerful example (Bolten and DeHaven, 1977). In particular, the Dutch water management domain was (and is) closely connected to some important water knowledge institutes, which are partly publicly funded, and to several academic departments specialised in civil engineering and water management. However, the horizontalisation that has characterised policymaking in many areas since the last decades of the 20th century, resulting in the tossing and turning over issues in multiple arenas between various actors, also had an impact on the role of policy analysis (focused on delivering authoritative facts that enable legitimate policy choices to be taken) in these areas and did not leave the water domain undisturbed. As conflicts arise over policies, the scientific foundations of these policies are no longer automatically taken for granted. As in other fields in the network-like policy settings, 33

Policy analysis in the Netherlands

battles of analysis are fought. In these battles, stakeholders try to convince each other by commissioning research and mobilising expertise. Policy advocacy results in a ‘rain’ of (unused) reports and contradictory truths. In this way, expertise and research contribute to uncertainty and deepen policy controversies rather than resolving them (Van Eeten, 1999). Recently, the rise of social media has eroded the authority of scientific institutions even further. Citizens seek information and evidence on the internet to support their preferences, and find diverging reports on evidence and knowledge claims of which the scientific tenability is uncertain (Klijn and Koppenjan, 2012). Since Dutch policy in general and water policy more particularly is renowned for its consensus nature, one might expect that practices have been developed to overcome these battles. Certainly, in knowledge-intensive areas like water management, one might expect developments in policy analysis that seek to find ways to make policy analysis authoritative, credible and robust in arenas of multi-actor decision-making and policy implementation. New approaches, such as joint fact-finding, knowledge co-production and negotiated knowledge, are developed to realise knowledge that is both credible (scientifically valid) and acceptable (socially robust) (Edelenbos et al, 2011; see also Chapter Twelve). In this chapter, a highly interesting case of joint fact-finding in Dutch water management is presented, in which some key dilemmas of knowledge co-production are nicely illustrated. This case – a planning process aimed at realising water retention in the Dutch polder of Reeuwijk in the western part of the Netherlands – shows the dilemmas of joint fact-finding when it comes to combining credibility and acceptability. The first round in this planning process was characterised by close cooperation between experts and bureaucrats, with stakeholders kept at a distance. In the second round, there was close collaboration between experts and stakeholders, and the policymakers were more at a distance. Both situations generated their own problems for realising ‘negotiated knowledge’.With this case study, we aim to illustrate to what extent practices have been developed in Dutch knowledge-intensive policy areas like water management that are aimed at enhancing the authoritativeness of (scientific) knowledge and expertise in networks in which the various actors that affect and are affected by government policies interact. Based on this sketch of the state of the art, we identify the challenges that both practitioners and scholars face in developing process-management strategies that aim at enhancing joint fact-finding between the triangle of experts, policymakers and stakeholders.

3.2: Knowledge for governance in complex networks: joint factfighting or joint fact-finding? We live in the era of governance (Pierre and Peters, 2000; Bingham et al, 2005), a development also characteristic of the water sector. The classical institutions of the state are supplemented by myriad temporal and structural governance networks in which mutually dependent actors and actor coalitions fight over a variety of objectives and problem perceptions. In these network settings, actors are not able to accomplish their objectives and to solve the problems they envision on their own. Governments, private business and stakeholders increasingly need to combine their resources and collaborate in order to co-produce solutions to wicked problems and deliver complex public services that meet the high demands of society. For governments, these 34

Policy analysis in networks: the battle of analysis and the potentials of joint fact-finding

developments require a shift from government towards governance: from traditional control and top-down steering towards negotiation and consensus building. However, these forms of collaboration and governance are hard to accomplish. Often, actors are not aware of their dependencies, or lack the skills to collaborate. Also, institutional structures and budgets may hinder collaboration. What is more, established forms of collaboration and existing networks may prevent new stakeholders from accessing the arena and addressing new problems that cut across existing boundaries, and prevent innovations from coming about (O’Toole, 1988; Gage and Mandell, 1990; Kickert et al, 1997; Agranoff and McQuire, 2001).Therefore, these collaborations often require management. This management may exist of spontaneous forms of facilitation and meditation by various actors. Also, conscious and planned attempts at the design and management of interaction processes among actors within networks may occur. In the literature on governance networks and network management, which, not by accident, has inspired and has been inspired by many policy scientists in the Netherlands, these activities are referred to as network management or meta-governance (Koppenjan and Klijn, 2004; Bell and Park, 2006; Sørenson and Torfing, 2007; De Bruijn and Ten Heuvelhof, 2008; Torenvliet et al, 2012). Meta-governance can be seen as the application of deliberate management strategies to facilitate interaction between actors and proven process-oriented norms like openness, participation, deliberation, fair play and equality to organise collaboration (McCreary et al, 2001;Thomson and Perry, 2006; De Bruijn et al, 2010). Within these governance network processes, the traditional role of scientific research and expertise has changed. It was long an undisputed fact that policy decisions had to be based on a scientifically proven, objective knowledge base (Lasswell, 1950; Simon, 1957). Certainly, it was considered necessary for identifying the most effective and efficient solutions to policy problems. It was for this reason that government agencies invested so much in generating policy-relevant information that could rationalise and improve public policymaking (Weiss, 1977; Stone, 2000). At the root of this positivistic belief was the assumption that it is possible to investigate policy problems exhaustively, and that science would therefore be able to investigate all relevant aspects of those problems, which would, in turn, facilitate informed decision-making. Policymakers within government have therefore been recruited on the basis on their substantive expertise regarding the issues that policies address. Within the Ministry of Transportation and Water Management, for instance, engineers have traditionally been employed. Scientific planning bureaus were established that were supposed to develop objective knowledge on problems and solutions on which policies could be based (see Chapter Nine). As many research activities were privatised or outsourced in due course, an extended knowledge industry of consultancy and engineering firms has evolved, supporting governments in designing and implementing their policies (Gibbons et al, 1994; In ’t Veld, 2010). However, science turned out to be an imperfect arbiter in societal controversies and a biased facilitator of rational choices in politics. It could not deliver the complete, undisputable and completely convincing underpinning of policy decisions. The rational-synoptic approach of policymaking does not reckon with the fact that many of the problems with which government is confronted are, by their nature, what are called ‘wicked’ problems (Rittle and Webber, 1973). Their complexity is not only caused by a lack of information, requiring research and solutions based on in-depth (scientific) knowledge, but also results from ambiguity: the fact that various actors that 35

Policy analysis in the Netherlands

affect or are affected by the problem or proposed policies have their own perception about the nature of the issue, its solutions and the values that are at play and should be taken into account.What is considered relevant by one actor may be seen as totally irrelevant by another. As a result, scientific research and expert knowledge are no longer automatically accepted in the arenas of network governance processes (Radin, 2002; Collins and Evans, 2007; Bijker et al, 2009; Hufen and Koppenjan, 2014). In addition to the presence of various values and perceptions in the policy arenas, the nature of the knowledge production process is also of importance in this respect. In contrast to the positivist idea that knowledge production is done in a neutral way, resulting in objective knowledge, it is increasingly acknowledged that knowledge is generated and mobilised in the context of a specific frame (In ‘t Veld, 2000; Nowotny et al, 2001). Moreover, such a frame gives rise to specific preferences with regard to methods of knowledge production (data collection, aggregation and interpretation). In controversial policy processes, actors with different frames commission their own research; they formulate partisan research questions based on their own interpretation of the relevant policy space and take seriously only that knowledge which corresponds with the main assumptions of their frame (Van Buuren, 2009; Van Turnhout et al, 2008). This often results in fact-fighting or a battle of analyses (Van Eeten, 1999;Van Buuren, 2009). In many situations of fact-fighting, scientists play the role of policy advocates (Pielke, 2007). They are part of an actor coalition in which they are closely related to public policymakers and/or stakeholders.The main boundary to be crossed is the boundary not between science and policy, but between different policy-advocacy coalitions that may have an institutional base in different arenas, networks or epistemological communities (Sabatier and Jenkins-Smith, 1993; Van Buuren and Edelenbos, 2004; see also Chapter Four). The fragmentation and contextualisation of the world of science (Nowotny et al, 2001) and stakeholders’ increased power to mobilise their own advocacy science have served to amplify stakeholders’ use of facts to defend their own and discredit opposing frames (Koppenjan and Klijn, 2004; Van Buuren and Edelenbos, 2004). No longer does one authoritative bureaucratic or scientific frame dominate; rather, there are multiple knowledge coalitions with their own frames that possess, mobilise and promote their own facts. Ignoring these facts when engaging in knowledge production may result in report wars, adding to uncertainty and conflict, rather than reducing it, thus endangering the potential to collaborate. Certainty is no longer found by mobilising an undisputed monopoly on science, but is constructed in a process of joint fact-finding in which each actor possess their own unique knowledge that is used as a resource to defend interests and positions (Jasanoff, 1994; Hoppe, 1999; De Bruijn et al, 2010).

3.3: The governance of knowledge production and knowledge use in governance networks Within the literature, many avenues are presented to prevent ‘wars of reports’ and to organise negotiated knowledge.What is needed for scientific research and expertise to become authoritative in the arenas of governance networks? In order to accomplish

36

Policy analysis in networks: the battle of analysis and the potentials of joint fact-finding

the involvement of research and experts, the literature suggests that the network process should meet the following demands (Koppenjan and Klijn 2004): 1. From a mandatory role or role as arbiter towards a facilitating role. Science and expertise needs to fulfil a facilitating rather than a mandating or arbitrating role. Rather than mandating scientists to design ready-made policies that only need to be implemented, experts and research may help to facilitate the discussion between actors in the arenas of governance networks to find appropriate processes of joint action (Salter, 1988). 2. Parallel organisation of research and negotiating.Traditionally, knowledge production is supposed to precede the process of policymaking and implementation. It results in evidence-based policy designs that have to be adopted and implemented by policymakers. However, as we argued earlier, since knowledge production is not neutral and policies are adopted in an arena of multiple actors with various perceptions, this will not result in authoritative knowledge. What is more, since the processes of knowledge production and policymaking have their own logics and rhythms, research often comes up with answers to questions not asked for, and at the wrong moments, leaving burning questions unanswered. In order to be able to perform a facilitating role, the activities of knowledge production and policymaking have to be organised in parallel instead of chronologically. In this way, knowledge questions that arise in the policymaking process can be introduced in the arena of knowledge production, and answers arrived at in this arena can be fed back into the arena of policymaking (De Bruijn and Ten Heuvelhof, 2003, 2008). 3. Joint commissioning. Since research always takes place under conditions of scarcity, choices have to be made regarding the issues addressed, the solutions included, the research questions, the assumptions, the research methods, the interpretations and the implications. If researchers decide on these choices by themselves, or follow the choices of the actors that commission the research, it is probable that the research findings will not be accepted by the other parties in the arena, who have their own perceptions and their own preferences. The joint commissioning and joint implementation of research may provide a way of overcoming this problem. By doing so, actors can negotiate about the choices made.This interaction process of knowledge production informs actors on the complexity of issues and the research task, contributes to the transparency of the knowledge production process, and commits them to the outcome of the research, even if these do not meet their prior expectations or preferences (Van Buuren et al, 2003). 4. Safeguarding boundaries. Joint commissioning and the parallel organisation of research activities and policymaking may result in the merger of the two arenas. If conflicts penetrate the arena of knowledge production, the added value of research and expertise will no longer be ensured. This is what actually happens in battles of analysis: if experts and researchers are engaged in processes of policy advocacy, their authoritativeness is at stake and their answers will contribute to uncertainty rather than solving it. Boundaries between science and policy demarcate the socially constructed and negotiated borders between science and policy and have important functions (eg protecting science from the biasing influence of politics) (Cash et al, 2002). However, these boundaries can become blurred, which means that science loses its legitimacy and credibility. Although boundaries can impede communication and hamper integration, they do have an important function in 37

Policy analysis in the Netherlands

guaranteeing authoritativeness. Therefore, dealing with the boundary between policy and science is ‘a balancing act’. To enhance the legitimacy of science in controversial multi-actor processes, the boundary between science and policy needs specific attention, which means attention to things like: selecting experts and researchers; providing an independent check on the process and outcomes of research by asking an independent forum of experts to comment; and safeguarding the transparency of the research process (Gieryn, 1983; Jasanoff, 1994). In the remainder of this chapter, we present a case study of a knowledge co-production process with specific attention to the question of how the four requirements just discussed played a role in this process and how this affected the outcomes of the process. In this chapter, a highly interesting case of joint fact-finding in Dutch water management is presented, in which some key dilemmas of knowledge co-production are nicely illustrated. This case concerns a planning process aimed at realising water retention in the Dutch polder of Reeuwijk in the western part of the Netherlands. This case study is a quite representative example of regional water governance (Van Buuren et al, 2010) and shows well the dynamics in the interaction between experts, policymakers and stakeholders in this knowledge-intensive but controversial policy domain. Actors involved in this knowledge-intensive policy area are, on the one hand, knowledge providers or producers (experts and researchers) and, on the other, knowledge consumers or recipients.Two categories of recipients can be distinguished. First of all, public policymakers (‘bureaucrats’) are important recipients and potential users of scientific knowledge and policy analyses. In the Dutch water domain, there is traditionally a strong relation (and thus less distance) between experts and bureaucrats. The second group of recipients are citizens and stakeholders, who are normally less strongly involved in policy processes. In order for knowledge to become authoritative, the involvement and interaction among these parties is required in a way that goes beyond the traditional dividing lines between knowledge production and knowledge use. We are interested in to what extent in this process of policymaking knowledge from research and experts succeeded in becoming authoritative and thus resulting in policies that are based on co-produced or negotiated knowledge, and to what extent the four requirements that were presented earlier hindered or enhanced this process.

3.4: Case analysis: enhancing retention capacity in the Dutch Gouwe Wiericke polder Methodology The research for this case was conducted between 2003 and 2006. Some 20 interviews were carried out with the main actors in this governance network (Van Buuren, 2006): representatives from the various public authorities (province, water board, municipality); various experts from the involved knowledge institutes; and representatives of the inhabitants (citizen and farmer associations). Furthermore, an extensive document analysis was conducted. The case study was analysed through the following three steps:

38

Policy analysis in networks: the battle of analysis and the potentials of joint fact-finding

1. The network governance process was reconstructed by distinguishing rounds of decision-making. These rounds are divided by crucial decisions that concluded earlier discussions and gave the discussions in the next round a new direction. More specifically, the role of (scientific) research and of experts within the interaction process was reconstructed. 2. Next we assessed to what extent the contribution of research and experts gained authoritativeness within the policymaking arena. Was scientific knowledge and expertise used, in the sense that co-production resulted in substantive solutions reflecting this scientific knowledge, linking it up with other forms of local knowledge? 3. Finally, we analysed to what extent the presence or absence of the four factors discussed earlier can explain why the (scientific) knowledge and role became authoritative in this case or not. To what extent did research and experts fulfil a facilitating role? Were research activities and policymaking linked in a parallel way? Were joint research activities employed? And were boundaries between knowledge production and policymaking safeguarded?

The case of Gouwe Wiericke The Dutch peat soil meadow areas have significant problems with their water management. Due to soil drop, water management is a difficult and expensive job in this area and low groundwater levels (which are required to enable agriculture in these areas) are difficult to realise, which has various negative consequences. Low water levels fasten soil drop and threaten the overall water quality, because of the intrusion of saltwater.

Round 1: retention areas as a formal policy objective for the Water Board (1998–2002) In the 1990s, these problems gave rise to the development of far-reaching proposals on the part of the Water Board Rijnland.This water management authority conducted research on the feasibility of water retention areas in the deep polders of Gouwe Wiericke West (a low-lying polder with serious soil drop problems) to solve a couple of water problems with regard to water quality (saltwater intrusion) and water storage. These studies, conducted by renowned consultancy bureaus like TNO and Arcadis, showed that water retention in this polder would be a ‘catch-all solution’ to the problems to do with water quality and quantity. In addition, these retention areas could also contribute to solving the problem of soil drop in the peat meadow areas of Rijnland.The outcomes of these studies were integrally incorporated in the policy plan of the Water Board and fitted nicely with their ambitions formulated in 1999 in the policy document ‘Bubbling waters’ (Provincie Zuid-Holland and Zuidhollandse Waterschapsbond, 1998). These studies also formed the starting point for a short but intensive decisionmaking process initiated by the Province of South-Holland and the Water Board of Rijnland in order to realise these retention areas.

39

Policy analysis in the Netherlands

Figure 3.1: Map of Gouwe Wiericke and De Venen

Round 2: towards decision-making on the realisation of a retention area (2001–04) An Environmental Impact Assessment (EIA) in 2003 was the first step in the formal planning process to ensure the realisation of these retention areas. However, unexpected process dynamics then appeared. When the EIA was launched during a public consultation meeting in March 2004, fierce resistance from the inhabitants of the polder was made public, as the general public absolutely disclaimed the knowledge disseminated about the problems in the polder. They felt the water problems to be overestimated and the qualities of the polder underestimated.The authorities felt that neglecting them would seriously complicate future discussions and the implementation of their ideas, and, thus, decided to form a collaborative Working Group made up of some of the angriest inhabitants. The Working Group was facilitated with the help of an independent public–private knowledge network (‘Habiforum’), which also delivered an independent process facilitator. In the first few months after its initiation, the Working Group was unable to do very much as the researchers tasked with conducting the EIA were busy working on their research, which was clearly demarcated by stringent requirements from governmental bodies. Stakeholders were not allowed to change the research focus of the EIA or to contribute to the research. Co-production was, in actual fact, not permitted. The mathematical and technical methods of the experts dominated knowledge production. After the installation of the Core Group, these models were supplemented with field visits and lay knowledge of the farmers from the polder, although their involvement was mainly symbolic. As a result, the stakeholders remained strongly disappointed with the extent of their input into the final EIA report. 40

Policy analysis in networks: the battle of analysis and the potentials of joint fact-finding

Ultimately, the EIA was not finalised at all. During the research process, it became undoubtedly clear that retention areas would be too expensive and would also generate too many risks in terms of water safety and soil stability. Therefore, the Water Board and the province decided to drop this option. At the same time, the responsible authorities understood that something had to be done to improve the economic conditions of the farmers in the polder. Therefore, they decided to give room for a new round of thinking and planning.

Round 3: the development of an alternative (2004–06) From this moment, a second round in the decision-making process was begun, at the initiative of the public authorities and Habiforum. They endorsed the necessity to give the inhabitants clarity about the future of the polder as soon as possible and to improve – if possible – the economic development options for the local farmers.The Working Group was given the option of delivering an alternative proposal, to find a solution to the water quality problems with more room for agriculture.The Working Group eagerly accepted this opportunity and within a few months, produced an initial rough sketch of the physical characteristics of the area and the main ambitions within the area with regard to agriculture, recreation, nature development, infrastructure and water quality. This sketch was the basis for more thorough analyses of the problems and potentials of the area, and formed the starting point of a search for an optimal mix of functions (based upon the ambitions of the regional authorities to improve water quality and local ambitions to improve living conditions). Experts and stakeholders worked together in the development of the Working Group proposal in joint field excursions and joint design sessions. A multi-criteria analysis (MCA) was used to judge the Core Group’s proposal and to compare it with two other scenarios. The MCA – which did not have many limitations placed upon it with regard to form and content – was able to bring together expert, bureaucratic and stakeholder knowledge. The resulting comparative and quantitative table was seen by all the actors involved as a good instrument for communicating the proposal put forward by the Core Group but did not convince the authorities and the bureaucrats who informed them. The officials and governors who were responsible for decision-making were only slightly included in the interactive process. They were unwilling to agree on this proposal because, in their eyes, it contributed too little to the realisation of their ambitions and they could not recognise their initial ambitions with regard to reducing the saltwater intrusion into other polder systems. The Water Board and province decided to implement some small elements of the proposal and postponed the ultimate decision over a new policy initiative: a Peat Meadow Contract for Gouwe Wiericke. This contract was formally agreed on in 2010 and has to be implemented in 2014.

3.5: Analysis: distance and co-production, shifting accents The authoritativeness of knowledge in the three rounds In the first round, knowledge became authoritative among experts and policymakers, but not among stakeholders. Due to the closed policy preparation, the considerations, ideas and local knowledge of stakeholders were not taken into account.The knowledge

41

Policy analysis in the Netherlands

of external experts fitted nicely with the ambitions of the regional policymakers and was directly transmitted into policy documents. This is slightly adjusted in the second round, when the EIA was conducted. In this round, there was some interaction between experts and stakeholders, but it was limited to some meetings and one site visit in which the farmers did not tell the experts the whole truth about, for example, the location of saltwater wells. At the end of this second policy round, the policy proposals were abandoned, not so much due to resistance, but due to the fact that they proved infeasible and costly: high ambitions had resulted in unfeasible solutions. The lack of input from other knowledge sources may be seen as a cause. In the third round, knowledge and solutions were co-produced among knowledge producers and stakeholders. Within this arena, participants succeeded in knowledge co-production. However, the co-produced solutions then entered the arena of formal decision-making. In this arena, the knowledge and the solutions that were based upon it were not authoritative. Overall, it can be concluded that actors within this policymaking process were not successful in producing knowledge in such a way that it supported the process of policymaking among the various actors involved and resulted in the acceptance of co-produced knowledge-based policy solutions.

The four requirements for gaining knowledge: authoritativeness as explanation Earlier, it became clear that knowledge developed in the various rounds did not succeed in becoming authoritative in the relevant arenas of decision-making: coproduction of knowledge failed. In this section, we investigate to what extent the four requirements for knowledge to become authoritative – facilitating, concurrent, co-commissioned knowledge production with proper boundary work – can explain for this.

1. Did knowledge fulfil a mandatory or facilitating role in the policy process? In the first round, scientific research had a mandatory role in policymaking. The process was aimed at designing a feasible solution to the problem and this design process was actually mandated to the experts that dominated the arena and the research activities they engaged in and controlled. This resulted in resistance on the side of the inhabitants of the polder. However, in the end, it was the experts’ inability to come up with affordable and feasible solutions that produced failure. Involvement of other actors and the concerns, considerations and local knowledge that they might have brought forward might have resulted in a less overoptimistic and more practical approach to the problems addressed. In the second round, knowledge procedures had a facilitating role with regard to the policymaking process, in which stakeholders had the lead. This contributed to the empowerment of these stakeholders. In the third round it became clear that the proposals developed in this way could not convince public officials that had not participated in the earlier process.The interaction among officials and stakeholders was not facilitated by this knowledge production.

42

Policy analysis in networks: the battle of analysis and the potentials of joint fact-finding

2. Did knowledge production relate to decision-making in a parallel or sequential way? In the first two rounds, knowledge production and policymaking were organised in a parallel way: the process of developing and accessing solutions had a strong technical nature in which experts and their research constantly provided input in a process of formulating decisions about how to solve the water problems of the polder. In the second round, the research and the development of solutions were likewise organised in a parallel way. Finally, an interative process between experts and the Working Group was organised in which ideas and knowledge were continuously exchanged and refined. 3.Was knowledge production co-commissioned? In all three of the rounds, co-commissioning limited itself to the boundaries between the various sets of knowledge recipients: policy officials on the one hand; stakeholders on the other. Within these coalitions of actors – the epistemic community of experts and policy officials from various agencies and layers of governments in the first round, and the participants in the collaborative process in the second round – joint commissioning occurred, but no joint commissioning occurred among these coalitions. As a result, the choices underlying this research were not jointly decided upon, thus reducing the chances of commitment of each of the parties to the outcomes of the knowledge production.

4. How were boundaries between knowledge and policy safeguarded? If the boundaries of the arenas of policymaking and knowledge production are blurred and both arenas merge, the independency of research is jeopardised, as is the chance that outcomes will be acceptable for parties not involved in these arenas. In the first round, it can be stated that this merger took place, given the close relationship between experts and policymakers. The process was dominated by technical arguments and the input of experts to such an extent that it can be said that the experts actually were the policymakers. In the second round, the role of experts was less dominant, but given the dominant role of stakeholders as commissioners or clients and the absence of other actors in the arena, it can be argued that boundary work between knowledge production and policymaking was limited and that little countervailing influences were present to protect the independent position of knowledge producers. As a result, in the third round, the role of the knowledge producers participating in the second round was compromised and limited, and their contribution had lost its authoritativeness. This analysis allows for at least two conclusions. First of all, the analysis shows that in this policymaking process, the four requirements were not or only partly met, thus providing explanations for the fact that the knowledge and the solutions that were based upon it did not become authoritative. Second, to a certain extent, some of the requirements were met, but only with regard to the linking up of knowledge producers with a subset of actors within a specific arena of the process. Applied in that way, they do not guarantee that knowledge production is linked up adequately to all relevant actors, coalitions and arenas within a policy process.

43

Policy analysis in the Netherlands

Table 3.1: Joint fact-finding in the case of Gouwe Wiericke Round 1

Round 2

Round 3

Mandatory or facilitating role?

Strictly mandatory

Mandatory

Facilitating towards stakeholders, not the interaction process

Parallel or sequential arrangement?

Parallel to policymakers; sequential to stakeholders

Parallel to policymakers; sequential to stakeholders

Parallel to stakeholders; sequential to policymakers

Joint commissioning or not?

No joint commissioning

No joint commissioning

No joint commissioning

Boundary work or not?

Arena merged

Arena merged

Limited boundary work

3.6: Explaining why knowledge and the solutions based upon it did not become authoritative What overall conclusion can be drawn regarding the explanation of why knowledge did not become authoritative in this process? An interesting alteration of roles in the process described earlier can be observed. In the first round, experts dominated the policy process and their knowledge was the main building block for making policy decisions by the Water Board Rijnland and the province. Stakeholder knowledge was neglected for a long time and stakeholders were locked out of the decision-making process. Even the installation of a forum for stakeholder participation during the process of conducting the EIA did not result in diminishing the distance between experts and citizens. In particular, the public officials from the Water Board used the same technical discourse as the experts, with a rather closed ‘epistemic community’ being created as a result (both belonged to the same expert community of geohydrologists and water management). As a result, citizens could not participate in this debate, even if they had access to the arena’s of policymaking, since they were not able to understand and use the technical arguments that were needed to influence the discourse.This closeness resulted in deep distrust among farmers and citizens, which was deepened by the cleavage between their local knowledge based upon empirical observation and the expert knowledge based upon formal models and mathematical calculations.The final EIA confirmed the opinion of the farmers that large retention basins in the polder were physically unfeasible. However, after the disappointing results of the EIA (for the authorities), stakeholders were given a much more central role in the process. They were allowed to develop their own proposal. Experts were given the role of critical reviewers and providers of supplementary data. A process of co-production was organised. This process was facilitated by an independent process-manager with much experience in controversial and complex planning processes. He was able to mediate between a variety of stakeholders and experts. Local knowledge was validated by experts from Wageningen University and was quantified for incorporation into the MCA as much as possible. At the same time, the public officials – in this round, mainly from the province – remained more distant from this interactive process. As a result, when officials in the traditional arenas of public decision-making had to decide upon the proposals developed by the Working Group, they were sceptical about the results from the MCA. They were not convinced of the effectiveness of the ‘Working Group’ proposal and gave negative advice to their principal, who ultimately decided not to implement it. 44

Policy analysis in networks: the battle of analysis and the potentials of joint fact-finding

Table 3.2: Co-production and distance in three rounds First round

Second round

Third round

Co-production

Between experts and policymakers

Between experts and policymakers

Between experts and stakeholders

Distance

Citizens/stakeholders not in the picture

Citizens/stakeholders at a distance

Policymakers at a distance

Main explanation

Epistemic community between experts with same disciplinary background

Epistemic community between experts with same disciplinary background

Dynamics of collaborative process: interaction only between involved actors. Officials in formal positions were not (enough) involved in former round

3.7: Conclusions: organising co-production and distance in the triangle In the Netherlands, knowledge-intensive policy domains like water management are traditionally characterised by strong (corporatist) relations between public authorities and knowledge institutes (Edelenbos et al, 2011). These relations are often rather closed to the engagement of other actors. Furthermore, experts and bureaucrats in these domains normally use the same language, which enhances the existing cleavage with ‘lay people’ and contributes to the distrust by farmers and citizens of the results of scientific expertise (see also Chapter Fifteen). This may jeopardise the feasibility of policy solutions in at least two ways. First, it may result in a lack of support from stakeholders that are affected by the solutions. Second, it prohibits the mobilisation and use of other sources of knowledge that may contribute to the quality and feasibility of the solutions considered. Our case study is an example of a larger development reflecting the acknowledgement of the need to deal with the network-like environment, even in knowledge-intensive policy areas. Problems can no longer be addressed by approaching them as solely an intellectual design process. Nor can they simply be imposed.Various stakeholders and their needs, perceptions and strategies need to be addressed.This has implications for processes of knowledge production and utilisation.A variety of new tools, techniques and practices have been developed to involve a wider set of stakeholders in problemsolving and policymaking. Given the knowledge-intensive nature of these policy areas, these new practices of process or network management also include organising processes of joint fact-finding between experts and stakeholders. At the same time, the case study also shows that despite the fact that the tools and methods of network management have become quite sophisticated, the ways in which they are applied are not successful in every way.The success of joint fact-finding can become its main risk: a strong sense of mutual belonging without critical reflection upon the question of how the outer world will react to the proposal that is developed (groupthink). Often, co-production between two clusters of the science–policy–society triangle becomes too close, with the third side being neglected. Within the literature on process and network practices in the Netherlands, ample examples can be found of successful collaboration and joint fact-finding within specific settings of policymakers, experts 45

Policy analysis in the Netherlands

and stakeholders resulting in joint outcomes that are rejected by established interest groups, policymakers and politicians within the traditional arenas of decision-making (Koppenjan, 2005;Van Buuren et al, 2007;Van Gils and Klijn, 2007). Joint fact-finding is thus essentially organising the interaction within the science– policy–society triangle. The involvement of all three parts is necessary to develop authoritative and legitimate knowledge that is able to contribute to decision-making about controversial issues. Developing a next generation of practices of network management that goes beyond the involvement of stakeholders and joint factfinding in policy development, and that succeeds in connecting the various actors in the whole network and bridging the diverging and often contradictory interests, perceptions and knowledge claims, is perhaps the main challenge that both theorists and practitioners working within the paradigm of network government in the Netherlands currently face.

References Agranoff, R. and McGuire, M. (2001) ‘Big questions in public network management research’, Journal of Public Administration Research and Theory, 11: 295–326. Bell, S. and Park, A. (2006) ‘The problematic metagovernance of networks: water reform in New South Wales’, Journal of Public Policy 26(1): 63. Bijker,W.E., Bal, R. and Hendriks, R. (2009) The paradox of scientific authority.The role of scientific advice in democracies, Cambridge, MA: The MIT Press. Bingham, L.B., Nabatchi, T. and O’Leary, R. (2005) ‘The new governance: practices and processes for stakeholder and citizen participation in the work of government’, Public administration review 65(5): 547–58. Bolten, J.H. and DeHaven, J.C. (1977) Protecting an estuary from floods – a policy analysis of the Oosterschelde, Rand. Cash, D., Clark,W.C., Alcock, F., Dickson, N., Eckley, N. and Jäger, J. (2002) ‘Salience, credibility, legitimacy and boundaries: linking research, assessment and decision making’, Research Programs, John F. Kennedy School of Government, Harvard University. Collins, H. and Evans, R. (2007) Rethinking expertise, Chicago, IL: The University of Chicago Press. De Bruijn, H. and Ten Heuvelhof, E. (2003) ‘Policy analysis and decision making in a network. How to improve the quality of analysis and the impact on decision making’, Impact Assessment and Project Appraisal 20: 232–42. De Bruijn, H. and Ten Heuvelhof, E. (2008) Management in networks. On multi-actor decision making, London and New York, NY: Routledge. De Bruijn, J.A., Ten Heuvelhof, E.F. and in ‘t Veld, R.J. (2010) Process management. Why project management fails in complex decision making processes, Dordrecht: Springer. Edelenbos, J.,Van Buuren, M.W. and Van Schie, N. (2011) ‘Co-producing knowledge: joint knowledge production between experts, bureaucrats and stakeholders in Dutch water management projects’, Environmental Science & Policy 14(6): 675–84. Gage, R.W. and Mandell, M.P. (eds) (1990) Strategies for managing intergovernmental policies and networks, New York, NY: Praeger. Gibbons, M., Limonges, C. Nowotny, H. Scott, S. and Trow, M. (1994) The new production of knowledge: the dynamics of science and research in contemporary societies, London: Sage. 46

Policy analysis in networks: the battle of analysis and the potentials of joint fact-finding

Gieryn,T.F. (1983) ‘Boundary-work and the demarcation of science from non-science: strains and interests in professional ideologies of scientists’, American Sociological Review 48(6): 781–95. Hendriks, F. and Toonen, T. (eds) (2001) Polder politics in the Netherlands; viscous state or model polity?, London: Ashgate. Hoppe, R. (1999) ‘Policy analysis, science and politics, from “speaking truth to power” to “making sense together”’, Science and Public Policy 26(3): 201–10. Hufen, J.A.M. and Koppenjan, J.F.M. (2014) ‘How evidence becomes authoritative in public policy implementation. Lessons from three Dutch white ravens’, Policy Studies 35(3): 264–81, DOI: 10.1080/01442872.2013.875148. In ‘t Veld, R.J. (ed) (2000) Willingly and knowingly. The roles of knowledge about nature and the environment in policy processes, Utrecht: Lemma. In ‘t Veld, R.J. (ed) (2010) Knowledge democracy. Consequences for science, politics, and media, Heidelberg: Springer. Jasanoff, S. (1994) The fifth branch. Science advisers as policy makers, Cambridge, MA, and London: Harvard University Press. Kickert, W.J.M., Klijn, E.H. and Koppenjan, J.F.M. (eds) (1997) Managing complex networks; strategies for the public sector, London: Sage. Klijn, E.H. and Koppenjan, J.F.M. (2012) ‘Governance network theory: past, present and future’, Policy & Politics 40(4): 187–206. Koppenjan, J.F.M. (2005) ‘The formation of public–private partnerships: lessons from nine transport infrastructure projects in The Netherlands’, Public Administration, 83(1): 135–57. Koppenjan, J.F.M. and Klijn E.H. (2004) Managing uncertainties in networks, London: Routledge. Lasswell, H.D. (1950 [1936]) Politics: who gets what, when, how?, New York. Lijphart, A. (1968) The politics of accommodation, Berkeley, CA: University of California Press. Mayer, I. (2007) ‘Evolution of policy analysis in the Netherlands’, in F. Fischer, G. Miller and M. Sidney (eds) Handbook of public policy analysis, Cambridge:Taylor and Francis, pp 553–71. McCreary, S.T., Gamman, J.K. and Brooks, B. (2001) ‘Refining and testing joint factfinding for environmental dispute resolution: ten years of success’, Conflict Resolution Quarterly 18(4): 329–48. Nowotny, H., Scott, P. and Gibbons, M. (2001) Rethinking science. Knowledge and the public in an age of uncertainty, Cambridge: Polity Press. O’Toole, L.J. (1988) ‘Strategies for intergovernmental management: Implementing programs in interorganizational networks’, Journal of Public Administration 11(4): 417–41. Pielke, R. (2007) The honest broker: making sense of science in policy and politics, Cambridge: Cambridge University Press. Pierre, J. and Peters, B.G. (2000) Governance, politics and the state, Houndmills: MacMillan. Provincie Zuid-Holland and Zuidhollandse Waterschapsbond (1998) ‘Bruisend Water. Perspectieven voor het waterbeheer’, Den Haag. Radin, B. (2002) Beyond Machiavelli – policy analysis comes of age, Washington, DC: Georgetown University 1101 Press. Rittel, H.W.J. and Webber, M.M. (1973) ‘Dilemmas in a general theory of planning’, Policy Sciences 4: 155–69. 47

Policy analysis in the Netherlands

Sabatier, P.A. and Jenkins-Smith, H.C. (1993) Policy change and learning. An advocacy coalition approach, Boulder, CO: Westview Press. Salter, L. (1988) Mandated sciences, Boston, MA: Kluwer Academic Publishers. Simon, H.A. (1957) Administrative behaviour: a study of decision-making processes in administrative organization, New York, NY: MacMillan. Sørenson, E. and Torfing, J. (eds) (2007) Theories of democratic network governance, Cheltenham: Edward Elgar. Stone, D. (2000) Policy paradox: the art of political decision making, New York, NY: W.W. Norton & Company. Thomson, A.M. and Perry, J.L. (2006) ‘Collaboration processes: inside the black box’, Public Administration Review 66(s1): 20–32. Torenvlied, R., Akkerman, A., Meier, K.J. and O’Toole Jr, L.J. (2012) ‘The multiple dimensions of managerial networking’, American Review of Public Administration 43(3): 251–72. Turnhout, E., Hisschemöller, M. and Eijsackers, H. (2008) ‘Science in Wadden Sea policy: from accommodation to advocacy’, Environmental Science & Policy 11(3): 227–39. Van Buuren, M.W, Klijn, E.H. and Koppenjan, J.F.M. (2003) ‘Dealing with wicked problems in networks: analysing an environmental debate from a network perspective’, Journal of Public Administration Research and Theory 13(2): 193–212. Van Buuren, M.W. (2006) Competente besluitvorming. Het management van meervoudige kennis in ruimtelijke ontwikkelingsprocessen, Utrecht: Lemma. Van Buuren, M.W. (2009) ‘Knowledge for governance, governance of knowledge. Inclusive knowledge management in collaborative governance processes’, International Public Management Journal 12(2): 208–35. Van Buuren, M.W. and Edelenbos, J. (2004) ‘Conflicting knowledge: why is knowledge production such a problem’, Science and Public Policy 31(4): 289–99. Van Buuren, M.W., Edelenbos, J. and Klijn, E.H. (2007) ‘Interactive governance in the Netherlands: the case of the Scheldt Estuary’, in M. Marcussen and J. Torfing (eds) Democratic network governance in Europe, Basingstoke: Palgrave Macmillan, pp 150–73. Van Buuren, M.W., Edelenbos, J. and Klijn, E.H. (2010) Gebiedsontwikkeling in woelig water. Over water governance bewegend tussen adaptief waterbeheer en ruimtelijke besluitvorming, Den Haag: Lemma. Van Eeten, M. (1999) Dialogues of the deaf: defining new agendas for environmental deadlocks, Delft: Eburon. Van Gils, M. and Klijn, E.H. (2007) ‘Complexity in decision making: the case of the Rotterdam harbour expansion. Connecting decisions, arenas and actors in spatial decision making’, Planning Theory & Practice 8(2): 139–59. Van Putten, J. (1982) ‘Policy styles in the Netherlands: negotiations and conflict’, in J.J. Richardson (ed) Policy styles in Western Europe, London:Allen and Unwin, pp 168–96. Weiss, C. (1977) Using social research in public policy making, Lexington, MA: LexingtonHealth.

48

FOUR

Patterns of science–policy interaction Robert Hoppe

4.1: Introduction: science-informed or expert policy advice in the Netherlands The traditional view casts policy analysis as advice on the authoritative choices that undergird public policies by public servants in a public bureaucracy to political authorities. Because this governmental, instrumental and cognitive conception provides an incomplete view (Radin, 2000), policy-related activities are now more broadly referred to as ‘policy work’ (Colebatch et al, 2010; Kohoutek et al, 2013).This concept captures a broader set of roles and activities in the making of public policies under conditions of networked governance.This chapter’s focus is on a special category of policy workers: people working as scientific experts for public policy in advisory bodies and/or knowledge centres. Their capacity and status as experts is precisely derived from their credentials and position in, or knowledge of, science. Qua experts, they are specifically tasked with translating or processing scientific evidence and thought into policy advice. This category of policy workers, apparently because of their scientific expertise, stays rather close to the traditional cognitive role of a policy analyst. However, different from all other types of policy workers, their role obliges them to act as ‘boundary workers’ between science and policy/politics. The Netherlands has many such expert policy advisers as boundary workers because the country has a highly developed, complex and diverse infrastructure for scienceinformed policy advice. Through the Legal Framework for Advisory Bodies (Kaderwet Adviescolleges) of 1996, the Dutch have legally regulated the frequent use of science1-informed policy advice by national government. The law stipulates rules for the establishment, composition, modus operandi and terms of reference of so-called permanent advisory bodies. These bodies have either formally strategic or technical-specialist functions. In addition, there are temporary advisory bodies, established for between four and a maximum of six years, devoted to politically salient but midterm issues. Finally, there are ad hoc or one-off commissions, devoted to a single issue (no yet covered by a permanent or temporary body), which require only a ministerial decree for their establishment. Apart from the Framework Law for advisory bodies, there are bureaucracy-related sites and arrangements for science-informed policy advice. First, there are the socalled departmental ‘knowledge chambers’. Second, since the 1990s, departments have increasingly used external consultancies for knowledge-based policy advice. Third, there are advisory bodies that fulfil the so-called ‘planning bureau function’, which are major providers of science-informed advice to the three basic areas of sustainable development (Profit, Planet and People) in government policies (see also Chapter Nine): the Centre for Economic Policy Analysis (Centraal Planbureau; 49

Policy analysis in the Netherlands

CPB), the Netherlands Environmental Assessment Agency (Planbureau voor de Leefomgeving; PBL)2 and the Netherlands Institute for Social Research (Sociaal en Cultureel Planbureau; SCP). These de facto independent and interdepartmental bureaus formally function as parts of departmental hierarchies of, respectively, the Departments of Economic Affairs, Environment and Infrastructure, and Social Affairs. Fourth, strategic and future-oriented science-informed advice on overall and longterm government policy is provided by the Scientific Council for Government Policy (Wetenschappelijke Raad voor het Regeringsbeleid; WRR), which is an independent advisory body with a special relationship to the prime minister’s Department of General Affairs. From this brief and non-exhaustive overview (and the in-depth treatment in other chapters of this book), it should be clear that the scope, complexity and diversity of the Dutch science-informed advice infrastructure beg the question: how do we make sense of the different sorts of policy advisers involved in a variety of science–politics interactions? Is there more ‘system’ than meets the eye in this kaleidoscope of transactions? Are there any overall trends discernible in the development of the policy advisory infrastructure? The next sections of this chapter deal with the following topics. First, a (comparative) multi-level framework for expert advice as boundary work will be briefly set out. Second, the analysis turns to the levels of boundary arrangements and organisations, as influenced by different types of policy politics and different policy domains with different types of problems. It will be shown that advisory boundary arrangements and organisations adapt to the average types of problems they deal with, and to the larger policy-political landscapes in which their advice is to be used. Third, the chapter describes how the level of political culture influences expert advice through ideologically preferred policy styles. Fourth, and finally, some future challenges for research and practice are highlighted.

4.2: A minimalist framework of expert policy advice as boundary work Policy analysis as advice to political authorities implies (science-informed) knowledge utilisation. Both natural and social scientists conceptualise this as knowledge transfer, dissemination, research use and impact on the policy process (Landry et al, 2001; Nutley et al, 2007).This models everyday clichés in scientific ways. Policymakers and politicians like to suggest that they are ‘on top’, and call on the services of scientists and experts as policy analysts, which are just ‘on tap’. Scientists and science-oriented policy experts continue to tell their myth about the independent smart guys who dare to speak ‘truth to power’. However, both ‘sacred’ narratives neglect the ‘profane’ truth of the two-way, interdependent character of knowledge production and communication between experts and policymakers (Bijker et al, 2009).The production of knowledge and policy advice cannot be fully described in terms of clear boundaries between science and politics. From a macro-sociological perspective, science–politics interactions are ongoing co-productions (Jasanoff, 2004) between the scientisation of politics and the politicisation of science (Weingart, 1983). Of course, at meso- and micro-levels, this does not mean a complete blurring, as in a seamless web. Given the need for participation from different institutional spheres, a division of work is called for. However, such a division is not easily settled or stabilised. 50

Patterns of science–policy interaction

To draw together usable insights from the older work on knowledge utilisation and more recent research perspectives on co-production through boundary work (Hoppe, 2005), this chapter uses a multi-level framework as a heuristic for understanding policy analysis and advice as science–politics interaction (Hoppe, 2010b). Boundary work can more formally be understood as attempts by actors to define practices in contrast to each other through demarcation, as well as to find productive coordination across these boundaries through a division of labour that is more or less stabilised (Halffman, 2003). Demarcation and coordination are two sides of the same coin. Concern for high-quality performance makes expert advisors and policymakers mutually dependent; yet, they have to guard their separate identities and formal independence. Therefore, boundary work is full of paradoxes and dilemmas; the relationship will always be contested. Boundary work can be depicted as science–policy interactions in a multi-level system. From a micro-agency perspective, science–politics boundary work is most clearly visible in specific policy advisory projects around particular topics. However, at a meso-organisational level, projects are carried out by boundary organisations or boundary arrangements, the wide variety of hybrid organisations that straddle and mediate the boundary between professional–academic networks and public sector or policy organisations. At a next level of policy-political analysis, such boundary organisations usually cluster around the typical problems in an issue or policy network. These problem-and-network structures display a particular type of policy politics, and, in turn, are permeated by a political-cultural sphere, the characteristics of which influence science–policy interfaces on all levels as policy styles.3 To present a comprehensive picture of the science–policy interface in policy analysis and advice by experts, therefore, means to understand multi-level science–policy interactions and the ways these levels interact and interpenetrate (see Figure 4.1).

Figure 4.1: Multi-level conceptual framework for understanding science–policy interactions Political cutural sphere Policy problem types

Professionalacademic networks

Policy network types

Boundary arrangements and (meta) organisations

Public/policy organisation

Boundary work (project level)



Impact/use (project level)

51

Policy analysis in the Netherlands

4.3: Boundary arrangements, organisations and policy politics in expert advice Boundary organisations and arrangements should be viewed as part of larger policy networks. Such networks do have policy issue politics, that is, the combination of types of cognitive processes (‘puzzling’) and the styles of competitive interaction (‘powering’) that are characteristic for policymaking in a particular domain (Hoppe, 2010a). It will be shown how a particular type of policy issue politics influences and constrains types of effective boundary organisations and arrangements (Hoppe, 2010b). Depending upon how political authorities, supported by policy-advising staff and scientific experts and/or consultants, frame or structure the problem, different governance styles will prevail (see Figure 4.2), and these will allow only certain types of boundary arrangements. In the case of structured problems (SP; a strong value consensus and informational certainty), a central-rational rule approach to governance permits ‘outsourcing’ problem solving to bureaucratic or scientific/professional, sometimes commercialised, closed epistemic (Haas, 1992) and technical communities. In the case of unstructured or ‘wicked’ problems (UP; high value dissent and lasting deep informational uncertainties), an agonistic governance style will come about, allowing numerous and different types of stakeholders to play fluid roles, perhaps with flexible boundary arrangements as spaces for open deliberation and social learning. The in-between problem types of moderately structured problems with value consensus but considerable informational uncertainty (MSPg), or with value dissensus but high informational certainty (MSPm), generate yet different styles of policy politics (see Figure 4.2).

Figure 4.2: Problem types and types of policy politics Far from certainty on required and available knowledge

UP

MSP(g)

Agenda-changing populist politics, agonistic participation, crisis management; deliberation and learning in emergent network(s)

Normal advocacy coalition politics and/or problem-driven search in policy subsystems

MSP(m)

SP

Transformative discourse coalition politics, accommodation strategies or conflict management in issue network(s)

Normal regulatory policy in professional/technical community

Close to certainty ... Far from agreement ... Source: Hoppe (2010a: 142, Figure 5.4).

52

Close to agreement on norms and values at stake

Patterns of science–policy interaction

Variation in expert advice as boundary work The introductory description shows an array of hybrid, boundary-crossing institutional forms for science–politics interaction. Explicitly established and institutionalised boundary organisations for expert advice are but one manifestation of a much broader ‘twilight zone’ of hybrid arrangements that straddle the institutional boundaries between the sciences and politics (Merkx, 2008; Hoppe, 2008a; Van Egmond and Zeiss, 2010). Based on an extensive literature review (Hoppe, 2005), and using Q Methodology to empirically assess the range of differences, seven types of discourse on boundary work capture the variety (Hoppe, 2008b): • Rational facilitation of political accommodation: a discourse used by experienced, prominent members of advisory bodies, or civil servants attending to look-out and knowledge functions in departmental agencies; they strongly believe in Dutch consensus-type democratic practices of flexibility and compromise; they feed this accommodation process with arguments derived from both sound science and knowledge rooted in ‘best practices’; and they facilitate orderly transgressions between science and politics in an atmosphere of mutual trust (Buevink and Den Hoed, 2007). • Knowledge brokerage: a discourse in use by civil servants or consultants (Bouwmeester, 2010; Schulz, 2010; De Wit, 2011), who, despite well-known cognitive impairments of politics and bureaucracy, and under favourable conditions, exploit opportunities for instrumental policy learning. • Mega-policy strategy: the typical discourse of expert advisors in government-oriented or commercial think-tank functions; they verify and critically examine long-term strategic policy guidelines and pivotal assumptions in policy beliefs-in-use in light of the most recent sound science and argument (Scholten, 2008; Den Hoed and Keizer, 2009). • Policy analysis ‘avant la lettre’: a discourse used by expert advisors working in the long-standing pragmatic relations and rules-of-the-game of established policy networks for, for example, financial-economic policy or health policy; they provide evidence-informed intelligence, that is, information derived from available and usable sound science (De Vries, 2008; Halffman, 2009: 60; Van Egmond, 2010; Den Butter, 2011). • Policy advice: in this type of discourse, advisors inside bureaucracies claim to span the boundary between policy analysts and ministers and top-level civil servants; they ‘advise the prince(s)’ about the acceptability and feasibility of policy proposals, incorporating (if possible) usable, best-available knowledge on ‘what works’. • Post-normal science advice: this is a discourse by expert advisors working on sustainability issues, inspired by ‘post-normal science’ and ‘extended peer review’, as advocated by Funtowicz and Ravetz (1992), and observed in practice by Hisschemöller et al (2001) and many others (eg Van der Sluijs et al, 2011;Turnpenny et al, 2011); they see issues of sustainability beset by so many uncertainties and value conflicts that ‘normal’ expert advice is obsolete; they wish to create and institutionalise more stable role and interaction patterns so that experts and policymakers may engage in (more) productive, open dialogue and integrated assessment of all (dis)advantages surrounding policy issues.

53

Policy analysis in the Netherlands

• Deliberative-procedural advice: this discourse is found in advisory bodies with civil servants as permanent secretaries, knowledge auditors checking ethical codes of conduct for expert advisers or designing hybrid forums for ‘wicked’ policy problems, and also some consultants (Merkx, 2008; Aarden, 2009; Bijker et al, 2009; Grijzen, 2010; Hoppe, 2010a: 146–50; De Wit, 2011); they cherish and foster highquality policy deliberation, which requires a clear and transparent procedure, and a set of process-criteria that allows robust but trusting parties, dissidents included, to fully and openly debate policy proposals and their concomitant uncertainties and normative struggles, each from their own perspective on the common good.4

Types of problems, types of expert advice The variety found may be linked to the different types of problems a political and administrative system has to deal with: domesticated or structured problems versus ‘wicked’ or unstructured problems, and their in-between types (see also WRR, 2006). Researchers routinely distinguish between technical or specialist, strategic, and ad hoc or temporary advisory bodies. By and large, these types fit the problem type they normally deal with. First, consider specialist and strongly technical advisory organisations that deliver instrumental, detailed, ready-for-implementation advice on largely domesticated problems, like the Royal Dutch Meteorological Institute (Koninklijk Nederlands Meteorologisch Instituut; KNMI) for weather forecasts and the hard-scientific aspects of climate change or the 1965-established Technical Advisory Commission for Dykes (Technische Adviescommissie voor de Waterkeringen;TAW), continued after 2005 as Expertise Network for Water Safety (Expertise Netwerk voor Waterveiligheid; ENW). Politicians and other government policymakers generally prefer usable or directly instrumental advice. Because political multitasking environments easily generate cognitive overload, and building political support needs unequivocal messages, they put a premium on simplification, and it is easy for them to delegate or outsource technical problem solving to (quasi-)autonomous agencies staffed by certified experts. Yet, instrumental advice and delegated technocracy as problem-processing arrangements only work well for domesticated problems. In addition, there are clearly specialist advisory bodies and knowledge institutes of broader scope and a non-technical nature – like the three planning bureaus (CPB, PBL and SCP), the Justice Department’s Research and Documentation Centre (Wetenschappelijk Onderzoek- en Documentatiecentrum; WODC) or the Health Council (Gezondheidsraad; Gr). Their main business is to: keep track of, and occasionally produce, sound science; develop future scenarios; and advise on appropriate, effective and efficient means for consensual goals.They are also frequently able to deliver instrumental, usable advice. In the Netherlands, the political consensus on goals is usually documented in considerable detail in a coalition agreement formally adopted in Parliament by the political parties constituting the government. Hence, such instrumental policy advice is welcome during the larger part of a short two- to three-year term of government.5 It appears that the larger part of experts in advisory functions operate like policy analysts, knowledge brokers or (in-house) policy advisors.They work the borderline between science and politics, dealing with domesticated problems or problems concerning appropriate policy instruments, or their zone of overlap. The Dutch planning bureaus 54

Patterns of science–policy interaction

(see next section) and civil servants heading the ‘knowledge chambers’ in Dutch national departments frequently represent this type of expertise. Next, consider strategic advisory institutes, with the WRR being the best example (see also Chapter Nine).6 This council deals with as yet undomesticated problems and politically sensitive issues characterised by potentially divisive ethical dilemmas. Examples are what to do with increasing amounts of labour migration to the Netherlands in the 1970s and later (Scholten, 2008; Van de Beek, 2010), or, already in the late 1970s, how to deal with environmental risk issues in economic policymaking (Hoppe, 1983). The WRR’s advice is usually conceptual and not immediately translatable into policy design, adoption or implementation. Rather, such advice has the character of future studies and scenarios, meta-policy arguments about major policy guidelines, and long-term policy. In addition, the council devises new policy paradigms and policy discourses in which protagonists and antagonists in ethically divisive issues may find opportunities for reconciliation, if only on a procedural or temporary basis. In these strategic advisory institutes, one naturally finds relatively more rational facilitators and mega-policy strategists. Compared to instrumental, short-term advice, strategic advice for the mid- and long term has a much narrower policy window. Timing advice well is both more important and more difficult: relatively propitious times appear to be just before, during and perhaps still in the first year after a change of government. However, short-term political developments often outpace strategic advice. A 2001 WRR study was an intellectually courageous attempt to redefine the policy paradigm for immigration and integration policy along transnational political trends and economic globalisation. Yet, published and presented briefly after 11 September 2001, the advice had become profoundly unwelcome in a political climate converting overnight to nationalistic assimilation (Scholten, 2008). Post-normalists and deliberative proceduralists are forced to operate across the borderline between domesticated and wicked problems. For example, an expert advisor working for the Rathenau Institute is supposed to inform Parliament, but indirectly departments and important stakeholders as well, on the newest technological developments and innovations and their ethical, legal and social consequences (Van Est et al, 2002;Van Est and Brom, 2012). Expert advisors in the Health Council, too, will be confronted by complex medical-technological issues with ethically unknown or divisive implications (Bijker et al, 2009). Hence, trained as medical specialists, they also need the skills to advise in debates on the ethical acceptability and goal appropriateness of medical-technological innovation (Hoppe, 2008a). The same goes for the advisors in ethical commissions for scientific research or in hospitals. Finally, in sustainabilityrelated issues, PBL advisors also need the skills to operate on all problem types. Because sustainability issues are far less stabilised in their division of labour between experts and policymakers, PBL advisors are frequently post-normalists or deliberative proceduralists (eg Hajer, 2012). Concluding and summarising this section, the links between types of policy advisory bodies – as more or less stabilised sets of boundary work discourses, rules and habits – and policy problem types may approximately be conceptualised as in Figure 4.3.

55

Policy analysis in the Netherlands

Figure 4.3: Problem types and some exemplary advisory bodies in the Netherlands Far from certainty on relevant, available knowledge

UP

Debates on means

Planning bureaus (esp. CPB, SCP) PBL WRR

WODC

Rathenau Health Council TAW/ENW

Ethics commissions

Close to certainty ...

Debates on ethics Far from agreement ...

KNMI SP

Close to agreement on norms and values at stake

4.4: Political cultures and policy styles in the Dutch advisory infrastructure Policy styles and public epistemologies At the policy-political level, boundary arrangements for expert policy advice cluster around specific (clusters of) policy problems and their typical issue or policy networks. These problem-and-network structures (Hoppe, 2010a) are, in turn, embedded in a political-cultural sphere (see Figure 4.1) (Halffman, 2005; Jasanoff, 2005; Engels et al, 2006; Lentsch and Weingart, 2009). From a political and policy science perspective, one may speak about policy styles (Richardson, 1982).The Dutch style is still frequently assessed as reactive-consensual, accommodating policy conflicts through lengthy negotiations, but leading to a complex and strong set of policy sectors, each with their own specialised governance networks (Van Putten, 1982; Skelcher et al, 2006; see also Chapter Three). Using a ‘Science, Technology and Society’ perspective and focusing on regulatory policy, authors like Renn (1995) and Halffman (2003, 2005) have enriched this view with the idea of regulatory regimes (Halffman, 2003: 360) or, in Jasanoff ’s (2005) terminology, ‘civic epistemologies’, that is, the cultural and organisational practices by which politically relevant knowledge is selected, filtered, deliberated, validated or challenged (Lentsch and Weingart, 2009: 7). They ‘intertextually’ permeate and penetrate science–policy interactions at all other levels; they are the ‘grand narratives’ on which types of knowledge and modes of expertise and advice giving in scienceinformed public policymaking are to be preferred. Three such ‘grand narratives’ of advice giving as boundary work are posited for the Netherlands: neo-corporatism, neoliberalism and a deliberative turn.

56

Patterns of science–policy interaction

Fluctuating neo-corporatism In neo-corporatist arrangements, a restricted set of a policy sector’s main actors or stakeholders are formally accredited to enter the arena for policymaking. Now weaker, now stronger, neo-corporatism has always been present in the Netherlands, mostly in socio-economic policymaking. In neo-corporatist policy networks, institutionalised expertise takes two forms. First, formally accredited stakeholder representatives mobilise their own expertise. In more technical negotiations, for example, on health insurance schemes, experts like university professors may even represent patients.This pattern dominated the old system of national advisory councils. In the second form, experts delineate the playing field for stakeholder negotiations. They wave a flag whenever the negotiation game exceeds known budgetary constraints or when, say, projections of next year’s economic growth become unrealistically high.This second pattern is to be found most prominently in the planning bureaus. The neo-corporatist tradition of ruling by consensus among an elite of relevant stakeholders – the model of recognised employer organisations and labour unions expanded to other sectors of society – led to an unchecked growth of the number of such councils. In the 1970s and 1980s, numerous reports advocated a reduction in their number (over 400 in 1976) and the creation of some order. Meanwhile, the nature of expertise in the councils professionalised, in the sense that mere representation of interests gradually shifted to interest-cum-knowledge representation (WRR, 1977; Oldersma, 2002). In 1997, the Ministry of the Interior both initiated the new framework law for advisory bodies and abolished nearly all of the existing ones.The Socio-Economic Council excepted – as an icon of the neo-corporatist tripartite consultation between labour, capital and state experts in socio-economic policy – all advisory councils were now considered expert councils. Also, councils were to become more generalist, covering more than one policy field. However, in practice, the logic of separate government departments with their own policy styles proved stronger than the logic of legal reorganisation (Klink, 2000). In 2009, there followed a second round of shrinking expert policy advice.Top-level civil servants and ministers desired to get rid of the excessive unsolicited, strategic advice emerging from the (modest) scope of advisory councils to set their own agendas (Hoppe, 2007). They successfully used the logic of performance measurement as a tool for eliminating some more advisory bodies (eg the Council for Spatial and Nature Research [Raad voor Ruimtelijk, Milieu- en Natuuronderzoek; RMNO]) and disciplining the remaining advisory councils and planning bureaus to deliver more actionable and instrumental advice. This time, one planning bureau, the National Bureau for Spatial Research (Ruimtelijk Planbureau; RPB), was eliminated and amalgamated with the PBL. This brings us to the second mode of institutionalised expertise under neocorporatism: the planning bureaus (Halffman, 2009). Different from suspicions raised by their being called ‘planning’ bureaus, these advisory bodies and knowledge institutes provide government departments with assessments of states of affairs and future developments in their policy sectors, and relate these to policy options.Although each of the three presently functioning planning bureaus (CPB, SCP, PBL) are formally agencies answering to departments whose ministers bear political responsibility for their finances and research agenda, through skilful performance of independence and political neutrality and fairness,‘they can provide policy makers with knowledge which is considered reliable and neutral to an extraordinary degree’ (Halffman, 2009: 41). 57

Policy analysis in the Netherlands

The central mission of the CPB is to discipline politics in terms of budgetary constraints and sound economic science. Here, independence clearly has its limits. The CPB, the oldest, most prestigious planning bureau, has close relationships with the civil servants of the Ministries of Economic Affairs and Financial Affairs, often with shared training in economics, econometrics or accountancy, and career patterns that switch between the department and planning bureau (Halffman, 2009: 48). The other planning bureaus were actually modelled after the CPB, as other departments mobilised their own knowledge and expertise. The environmental planning bureau, with experts from many different disciplines, is also engaged in the quantitative monitoring of policy instruments and impacts, and modelling the relationship between them. However, its mission is to monitor and remind the government of its commitments and promises regarding environmental policy. Where the CPB is willing to negotiate unexpected developments and resulting policy uncertainties with civil servants, and will never question intended government policy, the PBL (the former Milieu en Natuur Planbureau – MNP) experts insist much more on their scientific independence in defining uncertainties and are far less willing to uncritically accept government environmental policy (De Vries, 2008).The SCP is concerned with the description and assessment of current conditions in society, and less with projections of trends and future developments. Like the CPB, the SCP prefers numbers and calculations as inputs for its interpretive expertise about what is happening in Dutch society; its instrumental advice follows dominant policy frames without questioning them (eg Schnabel, 2000; Scholten, 2008).7 Despite their devotion to ‘serviceable truth’ (Jasanoff, 1990), the degree of acceptance by government and most policy actors of the planning bureaus’ assessments as, for all practical purposes, ‘uncontested’ is remarkable. Even political parties now routinely submit their election manifestos to the CPB and PBL for a calculation (‘doorrekenen’) and assessment of the likely outcomes.These are reported to the media and the voters as school report marks (Huitema, 2004). As such, planning bureaus occupy positions of obligatory passage points for Dutch politics that would be considered unacceptably technocratic in most other democratic countries. There appears to be a contradiction here. In spite of being ‘on tap’ to policymakers in neo-corporatist governance, it frequently looks like the planning bureaus are ‘on top’: they rationalise political debate, enforce budgetary constraints (even on political parties before elections) and contribute to the transparent accountability of the government’s policy performance. They manage to do this because of a number of reasons. They are in a privileged position regarding access to (government) data and calculative resources (user-oriented modelling; eg Petersen, 2006; Van Egmond and Zeiss, 2010). They successfully exploit the image and rhetoric of numbers as objective and neutral. They not only discipline the policymakers, but are also mobilised by the policymakers themselves to discipline each other.Their most important political function may well be in creating a shared definition of reality, without which accommodation of policy conflicts through negotiation would be near impossible (Halffman, 2009: 54). In other words, the advisory councils and especially the planning bureaus are a bulwark against ‘fact-free politics’. They turn even unstructured problems into structured or moderately structured ones with presupposed value and goal consent. Both in the advisory councils and through the planning bureaus, the neocorporatist logic of interest-cum-knowledge representation has changed into one of representation of the issues and the state of ‘relevant’ knowledge.This does not mean 58

Patterns of science–policy interaction

that the barriers to policy access to just a handful of major policy players, typical for neo-corporatist politics, have disappeared. Rather, experts have been repositioned in such a way that the executive has stronger leverage to break through corporatist deadlocks (Hemerijck, 1994).

Stable neoliberalism and highly distributed expertise The development of expertise as linesmen in neo-corporatist governance networks is only one trend. Another development clearly shows stable entrenchment of a neoliberal pattern. Typical for this pattern is a small state philosophy, doing more with less, leading to the externalisation of the in-house expertise of government departments, and an emphasis on the market coordination of expert resources. Here, two more indications for the stabilisation of neoliberal expert arrangements will be mentioned: the externalisation of expertise at all ministries, and the contractualisation and commodification of expert knowledge (see also Chapter Ten). It is in the context of internal departmental organisation of expertise and external relations with expert organisations that neoliberal solutions are most manifest. After a decade of confusion over how to organise for ‘knowledge-intensive administration’ or ‘evidence-informed policy’, all government departments have moved towards socalled ‘knowledge chambers’,8 sometimes with coordinating ‘scientific officers’ or ‘chief scientists’.9 Typically, the function of knowledge chambers is formulated as a quasi-market, that is,‘direct interaction between the supply and demand of knowledge’. Demand is represented by top-level departmental bureaucrats; supply by the leadership of knowledge institutes. Usually, annual consultations result in ‘strategic knowledge agendas’ that define long- to mid-term policy and research themes. The presumably improved connection between knowledge demand and supply would enable departments a larger and more complete overview of available expertise, and to draw on the just-right type of expertise per policy theme. In this way, departments are no longer the addressees of coherent expert advisory reports by councils.The externalisation of expertise contributes to more instrumental advice, to be selected by departmental civil servants, despite the longer- and mid-term policyand-research themes in the no-longer independently drawn-up strategic policy agenda. Yet, paradoxically, the knowledge centres and chambers present an image of a larger distance between science and politics to the outside world, and this may create a resource for building political legitimacy by asserting the ‘independence’ of expert advice. However, it also creates a gap between immediate policy needs and the agenda of professional researchers.This induces complex and increasingly formalised negotiations over mutual relations and degrees of control over research agendas (eg Van Hoesel and De Koning, 2005: 122–5). Another form of externalisation is the increased use of outside consultants, usually organised in a small number of large (>100 employees) commercial research and consultancy organisations like McKinsey, Berenschot, Ecorys or KPMG, and several dozens medium (10–100) and small (