The Technocratic Challenge to Democracy 2019052665, 2019052666, 9780367358280, 9780429342165

Citation preview

Routledge Research on Social and Political Elites

THE TECHNOCRATIC CHALLENGE TO DEMOCRACY Edited by Eri Bertsou and Daniele Caramani

The Technocratic Challenge to Democracy

This book represents the first comprehensive study of how technocracy ­currently challenges representative democracy and asks how technocratic politics undermines democratic legitimacy. How strong is its challenge to democratic institutions? The  book offers a solid theory and conceptualization of technocratic politics and the technocratic challenge is analyzed empirically at all levels of the national and supra-national institutions and actors, such as cabinets, parties, the EU, independent bodies, central banks and direct democratic campaigns in a comparative and policy perspective. It takes an in-depth analysis addressing elitism, meritocracy, de-politicization, efficiency, neutrality, reliance on science and distrust toward party politics and ideologies, and their impact when pitched against democratic responsiveness, accountability, citizens’ input and pluralist competition. In the current crisis of democracy, this book assesses the effects of the technocratic critique against representative institutions, which are perceived to be unable to deal with complex and global problems. It analyzes demands for competent and responsible policy making in combination with the simultaneous populist resistance to experts. The book will be of key interest to scholars and students of comparative politics, political theory, policy analysis, multi-level governance as well as practitioners working in bureaucracies, media, think-tanks and policy making. Eri Bertsouis Postdoctoral Researcher in Comparative Politics at the University of Zurich, Switzerland. Daniele Caramaniis Professor of Comparative Politics at the University of Zurich, Switzerland.

Routledge Research on Social and Political Elites Who are the elites that run the world? This series of books analyses who the elites are, how they rise and fall, the networks in which they operate and the effects they have on our lives. Series Editors: Keith Dowding and Patrick Dumont, Australian National University, Australia.

Ministerial Survival during Political and Cabinet Change Foreign Affairs, Diplomacy and War Alejandro Quiroz Flores Government Formation and Minister Turnover in Presidential Cabinets Comparative Analysis in the Americas Edited by Marcelo Camerlo and Cecilia Martínez-Gallardo The Selection of Politicians in Times of Crisis Edited by Xavier Coller, Guillermo Cordero, and Antonio M. Jaime-Castillo National Political Elites, European Integration and the Eurozone Crisis Edited by Nicolò Conti, Borbála Göncz and José Real-Dato Semi-presidentialism, Parliamentarism and Presidents Presidential Politics in Central Europe Miloš Brunclík and Michal Kubát The Contested Status of Political Elites At the Crossroads Edited by Lars Vogel, Ronald Gebauer and Axel Salheiser The Technocratic Challenge to Democracy Edited by Eri Bertsou and Daniele Caramani

For more information about this series, please visit: https://www.routledge.com/ Routledge-Research-on-Social-and-Political-Elites/book-series/RRSPE

The Technocratic Challenge to Democracy Edited by

Eri Bertsou and Daniele Caramani

First published 2020 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 52 Vanderbilt Avenue, New York, NY 10017 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2020 selection and editorial matter, Eri Bertsou and Daniele Caramani; individual chapters, the contributors The right of Eri Bertsou and Daniele Caramani to be identified as the author of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. 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 Names: Bertsou, Eri, editor. | Caramani, Daniele, 1968- editor. Title: The technocratic challenge to democracy / edited by Eri Bertsou and Daniele Caramani. Description: Abingdon, Oxon ; New York, NY : Routledge, 2020. | Includes bibliographical references and index. Identifiers: LCCN 2019052665 (print) | LCCN 2019052666 (ebook) | ISBN 9780367358280 (hardback) | ISBN 9780429342165 (ebook) Subjects: LCSH: Democracy. | Technocracy. | Technology and state. Classification: LCC JC423 .T347 2020 (print) | LCC JC423 (ebook) | DDC 321.8--dc23 LC record available at https://lccn.loc.gov/2019052665 LC ebook record available at https://lccn.loc.gov/2019052666 ISBN: 978-0-367-35828-0 (hbk) ISBN: 978-0-429-34216-5 (ebk) Typeset in Times New Roman by Lumina Datamatics Limited

Contents

List of figures List of tables List of contributors Preface and acknowledgements Abbreviations Introduction: the technocratic challenge to democracy

vii viii x xiv xvii 1

DANIELE CARAMANI

PART I

Concepts and theory

27

1 Technocracy and political theory

29

CHRISTOPHER BICKERTON AND CARLO INVERNIZZI ACCETTI

2 Neoliberal technocracy: the challenge to democratic self-government

44

IGNACIO SÁNCHEZ-CUENCA

3 Technocracy and depoliticization

61

PIER DOMENICO TORTOLA

4 Technocratic responsiveness

75

REINOUT VAN DER VEER

5 Measuring technocracy ERI BERTSOU AND DANIELE CARAMANI

91

vi  Contents PART II

Institutions, actors and policies

111

6 Technocratic cabinets

113

MARCO VALBRUZZI

7 Technocrats in cabinets and their policy effects

131

DESPINA ALEXIADOU

8 Technocratic cabinets in European negotiations

148

SILVANA TÂRLEA AND STEFANIE BAILER

9 Technocracy in discourse: the case of direct democracy

164

JEAN NAVA, MIGUEL ÁNGEL CENTENO AND LARRY LIU

10 Technocracy and the policy process

183

CLAIRE A. DUNLOP AND CLAUDIO M. RADAELLI

PART III

Comparative perspectives

197

11 The EU between technocratic and democratic legitimacy

199

MARINA COSTA LOBO AND IAN MCMANUS

12 Technocracy in Latin America: between stability and democratic deficit

216

EDUARDO DARGENT

13 Technocracy in Central/Eastern Europe and its impact on democratization

232

JOSHUA A. TUCKER AND JAN ZILINSKY

Conclusion—Technocracy and democracy: friends or foes?

247

ERI BERTSOU

Appendix References Index

271 279 312

Figures

I.1 The triangular relationship between representative democracy, technocracy and populism 4.1 Strategic interaction between a technocratic executive and its environment 5.1 Attitudes towards experts in governance 5.2 Partyness of government in Western Europe, 1945–2012 5.3 Central bank independence: Index of Central Bank Independence in OECD (Organisation for Economic Co-operation and Development) countries 6.1 Graphical configurations of the concept of technocracy as a “radial” category 6.2 Technocraticness of government in EU-28 member states (%), 1945–2018 7.1 Professional background of finance and social affairs ministers 7.2 Predicted number of economic reforms by technocrat finance ministers 9.1 Three-dimensional typology for regimes 9.2 Number of news updates by month for both campaigns 9.3 Topic prevalence and top topic terms in corpus 9.4 Difference in topic prevalence by campaign: Stronger in Europe minus Vote Leave 9.5 Word choice by campaign for “Groups Backing Campaign” Topic 9.6 World choice by campaign for “Cost” topic 9.7 Word choice by campaign for “Movement” topic 9.8 Difference in topical content for Topics 19 and 20 10.1 Expanding the epistemic mode 10.2 Conceptualizing knowledge modes as policy learning 11.1 “The EU or the IMF are the most effective for solving the economic and financial crisis”: percentage who agree (EMU average vs. GIIPS average, 2009−13)

20 81 102 104 106 115 128 138 143 168 171 173 174 176 177 178 179 188 192 207

Tables

3.1 5.1 5.2 5.3 6.1 6.2 6.3 6.4 6.5 7.1 7.2 7.3 7.4 7.5 7.6 8.1 8.2 8.3 8.4 9.1 10.1 10.2 11.1 11.2 11.3

Technocracy and depoliticization: a unified conceptual framework A typology of research in technocratic politics Classification of elite profiles Codebook for the analysis of technocratic discourse (excerpt) A typology of “technocratic categories” Technocrat-led governments in EU-28 member states, 1945–2018 Non-caretaker technocrat-led governments and their policies A typology of technocrat-led governments Types of technocrat-led governments in EU-28 member states, 1945–2018 Frequency of technocrats and experts in the portfolios of finance and social affairs (1980–2012) The role of technocrat finance ministers in economic reforms The role of social affairs ministers in social welfare reforms The role of finance ministers in public social spending, old-age pensions and health The role of expert finance ministers in public social spending, old-age pensions and health The role of experienced finance ministers in public social spending, old-age pensions and health List of technocratic governments during the Eurocrisis years Average government position during EMU reform negotiations 2010–15 Rejections and formal statements in the EU Council of Ministers Unpopular reforms after a technocratic government came to power Topics discussed in depth Degenerations of epistemic learning Possible roles of experts Support for technocracy as crisis manager in the Eurozone (2009–14) GIIPS country models Centre and left effects in countries with governmental bailouts (2010, 2014)

73 95 97 99 117 120 124 125 127 139 141 142 144 145 146 149 154 156 158 175 189 193 206 209 212

Tables  ix 13.1 Main results: Government evaluation and experience with technocracy 13.2 Views about government performance improvement 13.3 Openness to authoritarianism (LiTS data) A.1 List of issues (Chapter 8) A.2 Topic model results: top 20 terms by estimated probability, all topics (Chapter 9) A.3 Full model specification of Table 13.1: Government evaluation and experience with technocracy (Chapter 13) A.4 Full model specification of Table 13.2: Views about government performance improvement (Chapter 13)

240 241 242 271 273 276 277

Contributors

Despina Alexiadouis Chancellor’s Fellow at the University of Strathclyde, Glasgow. She is the author of Ideologues, Partisans and Loyalists: Ministers and Policymaking in Parliamentary Cabinets (2016). She  has published articles on technocracy in the Oxford Research Encyclopaedia and in the European Journal of Political Research. Stefanie Baileris Professor of Political Science at the University of Basel. Her research interests encompass decision making at the European and international levels, parliamentarians and parliamentary careers in Western European parliaments, and negotiations in the EU and international organizations. Her work has been published in various scholarly journals. Eri Bertsouis Postdoctoral Researcher in Comparative Politics at the University of Zurich. She published on citizens’ technocratic attitudes in West European Politics (2017) and American Journal of Political Science (forthcoming). Her research focuses on comparative political behaviour in Europe, specifically citizen attitudes towards democracy and technocracy, political trust, representation and partisanship. Christopher Bickertonis Reader in Modern European Politics and a Fellow of Queens’ College at Cambridge University. He is the author of European Union Foreign Policy: From Effectiveness to Functionality (2011), European Integration: From Nation-States to Member States (2012) and The European Union: A Citizen’s Guide (2016). On technocracy, he has written in the Oxford Handbook on Populism (2017) and the Critical Review of International Social and Political Philosophy (2015). Daniele Caramaniis Professor of Comparative Politics at the University of Zurich. He is Co-Director of the Constituency-Level Elections Archive (CLEA), the author of The Nationalization of Politics (2004) and The Europeanization of Politics (2015), and the editor of the textbook Comparative Politics (2020, fifth edn). On technocracy, he has authored an article in the American Political Science Review (2017) and a forthcoming article in the American Journal of Political Science.

Contributors  xi Miguel Ángel Centenois Musgrave Professor of Sociology at Princeton University. He  is the author of Democracy within Reason: Technocratic Revolution in Mexico (1994), editor of The Politics of Expertise in Latin America (1998) and author of several articles on technocracy and Latin America in various scholarly journals. Marina Costa Lobois Principal Researcher at the Institute of Social Sciences of the University of Lisbon. She is Director of the Observatory of the Quality of Democracy at ICS-UL and Vice-Director of Instituto de Políticas Públicas. Currently, she is Principal Investigator for the ERC Consolidator Project MAPLE, on politicization of the Eurozone crisis. She was one of the founding directors of the Portuguese Election Study project. Her research interests include the role of leaders in electoral behaviour, economic voting, political parties and institutions. Her latest book, Personality Politics: Leaders and Democratic Elections (2015), was co-edited with John Curtice. Eduardo Dargentis Associate Professor of Political Science at Pontificia Universidad Católica del Perú, Lima. His main teaching and research interests are comparative public policy, and democratization and the state in the developing world. He has published in various scholarly journals. His book Technocracy and Democracy in Latin America was published in 2015. Claire A. Dunlopis Professor of Politics at the University of Exeter. A public policy and administration scholar, her main research fields include the politics of expertise and knowledge utilization, epistemic communities and advisory politics, risk governance, policy learning and analysis, impact assessment and policy narratives. Her recent co-edited volume (with Claudio M. Radaelli and Philipp Trein) is Learning in Public Policy: Analysis, Modes and Outcome (2018). Her work has been published in various scholarly journals, and she is editor of the journal Public Policy and Administration. Carlo Invernizzi Accettiis Associate Professor of Political Science at the City University of New York (City College) and Associate Researcher at the Center for European Studies of the Institut d’Etudes Politiques de Paris (Sciences Po). He is the author of Relativism and Religion (2015) and What Is Christian Democracy? Politics, Religion, Ideology (2019). On technocracy, he has co-authored articles in the Oxford Handbook on Populism (2017) and the Critical Review of International Social and Political Philosophy (2015). Larry Liuis a Ph.D. candidate in the Department of Sociology at Princeton University. The co-authored chapter in this volume is his first publication. Ian McManusis Assistant Professor of Political Science at Marlboro College. He  was formerly an LSE Fellow in Social Policy at the London School of Economics and Political Science, a Postdoctoral Research Fellow at the University of Lisbon and a recipient of a German Academic Exchange Service doctoral research grant (Otto Suhr Institute of Political Science at the Free

xii  Contributors University of Berlin). His research interests include welfare state politics, social inequality, political parties and institutions, political polarization and public opinion. Jean Navais a Ph.D. candidate at Princeton University and currently works as a data scientist. His research focuses on discourse, narratives, sentiment analysis and political economy. He has conducted research on developmental institutions in Portugal and Latin America, and on the US economic policy, focusing on the Federal Reserve System after World War II. Claudio M. Radaelliis Professor of Public Policy at University College London. He  has published on the politics of evidence-based policy, governance, policy learning and technocracy in the European Union, including Learning in Public Policy (co-edited with Claire A. Dunlop and Philipp Trein; 2018) and Handbook of Regulatory Impact Assessment (co-edited with Claire A. Dunlop; 2017). Ignacio Sánchez-Cuencais Professor of Political Science at Carlos III University, Madrid, and Director of the Carlos III-Juan March Institute of Social Sciences. He is the author of The Historical Roots of Political Violence (2019) and co-editor of Controlling Governments (2008), as well as various articles in scholarly journals. On technocracy, he published an article on EU technocracy in the Annual Review of Political Science (2017). Silvana Târleais Postdoctoral Researcher at the Institute for European Global Studies and the Department of Political Science of the University of Basel. Previously, she was a Max Weber Fellow at the European University Institute, Florence, after obtaining her Ph.D. from Nuffield College, Oxford University. Her work on the European financial crisis and on the political economy of education has been published in various scholarly journals. Pier Domenico Tortolais Assistant Professor of European Politics and Society at the University of Groningen. He  recently co-edited Governing Europe: How to Make the EU More Efficient and Democratic (2017) and published various articles in scholarly journals. On the topic of technocracy, his interests revolve around the role and politicization of the European Central Bank during the Eurocrisis. His latest work on this subject has appeared in the European Journal of Political Research and the Journal of Common Market Studies. Joshua A. Tuckeris Professor of Politics, and Affiliated Professor of Russian and Slavic Studies, and of Data Science, at New  York University. He is a co-founder and co-director of the NYU Center for Social Media and Politics (csmapnyu.org) and SMaPP laboratory, the Director of NYU’s Jordan Center for Advanced Study of Russia, and co-editor of the award-winning blog “The Monkey Cage” at The Washington Post. His research has appeared in more than two dozen scholarly journals, and he is a co-author of Communism’s Shadow (2017).

Contributors  xiii Marco Valbruzziis Assistant Professor at the University of Naples - Federico II. He  is the author of A Changing Republic: Politics and Democracy in Italy (2015), as well as two books in Italian. On technocracy, he has co-authored two articles in the European Journal of Political Research (2014) and the Journal of Modern Italian Studies (2012). Reinout van der Veeris a Ph.D. candidate in Political Science at Erasmus University, Rotterdam. His research focuses on the responsiveness of insulated, technocratic executives to their wider political context, with a specific interest in the European Union, its institutions, and their relationship to public opposition. His work on technocracy and the EU has appeared in European Union Politics (2018) and the Journal of European Public Policy (2018). Jan Zilinskyis a Ph.D. candidate in the Department of Politics at New  York University and Research Associate at the NYU Social Media and Political Participation (SMaPP) laboratory. His research focuses on political knowledge, voter learning and mass political behaviour. He has co-authored an article in the Journal of Comparative Economics (2016).

Preface and acknowledgements

Recent manifestations of technocracy include “war” declarations from Mexico’s new president on the technocrats from the Salinas revolution in the 1990s, but also recent technocratic cabinets in Bulgaria, the Czech Republic, Greece, Hungary and Italy; non-partisan ministers in Portugal; and the appointment of new versions of neoliberal “Chicago Boys” in President’s Bolsonaro cabinet in Brazil. More generally, one observes the growth of supranational technocratic bodies such as the International Monetary Fund or the European Union. Examples also include populist attacks against the independence of the Federal Reserve and the European Central Bank (as aired by US President Trump or Greek Finance Minister Yanis Varoufakis, among others), pension institutions (by former Interior Minister Matteo Salvini in Italy), and courts and judges (by tabloids in Britain and populists in Switzerland), as well as the Yellow Vests’ protest against Emmanuel Macron, the French president often described as technocratic. Brexit Minister David Davis and the UK Independence Party have accused the civil service of sabotaging Britain’s exit from the European Union. Similarly, it is claimed that climate scientists and experts act based on “an agenda”, and that they are politicized and not neutral. On the opposite side, the reliance on experts finds increasing support driven by citizens’ scepticism towards bickering parties and politicians, and by distrust towards democratic institutions’ efficiency and competence in an increasingly complex and interconnected world. Issues of technocratic neutrality, its politicization, its role as counterweight to populism, the role it plays in responsible and responsive representation in democratic systems, but also the “democratic deficit” of technocratic regimes are the themes of this volume. Although the issue of technocratic politics is propelled into the public debate when technocratic forces manifest themselves, it has become clear that the questions it poses have roots in the very foundations of our governance systems and will become ever more pressing in the decades to come. As the complexity of political systems increases, due to technological advances and an interconnected world, and citizen demands for efficient outcomes grow, the tension between responsible and responsive governance will intensify. How can democratic systems manage to use independent knowledge and expertise to deliver effective governance without losing their democratic credentials? While the “technocratization” of politics (decision making being transferred to

Preface and acknowledgements  xv unresponsive, unelected elites) is often considered the underlying reason for the current populist backlash, technocracy can also offer a corrective for democratic systems that swing too far towards over-responsive governance. We therefore see technocracy as a challenge, but also as a potential corrective force, as a “friend” and as a “foe” of democracy. The aim of this volume is to understand and explain these dynamics, both in theory and in practice, and to provide a common framework for the study of technocratic politics for the future. This book developed out of various research initiatives. In 2006, the University of Zurich launched a broad research programme on the challenges to democracy in the twenty-first century funded by the Swiss National Science Foundation (NCCR programme). While dealing primarily with populism and mediatization, parts of its research were devoted to the growing demands for expertise and supranational governance in the context of globalization. At the closing of the ­programme in 2017, it was clear that beside populism, technocratic governance constituted an equally challenging alternative to representative democracy, albeit a neglected one. This  prompted theoretical work (Caramani 2017) and, eventually, empirical research on technocracy, most notably with a new comparative survey on technocratic attitudes among European publics (Bertsou and Caramani 2017). In  October  2017, the new research cluster on technocracy at the chair of Comparative Politics in Zurich organized a two-day workshop on “The Technocratic Challenge to Democracy”, followed by a workshop at the ECPR (European Consortium for Political Research) Joint Sessions of Workshops in Nicosia in 2018 and a book panel at the Council for European Studies in Madrid in 2019. For these workshops, leading academics on technocracy and related subjects were able to come together for the first time and think systematically from comparative politics and public policy perspectives about technocracy in the current “crisis” of representative democracy. This book is the result of these efforts. Our thanks go to Hanspeter Kriesi, who launched and directed for most of its history the NCCR research programme. We are grateful to the University of Zurich for funding the first workshop in 2017. Thanks go to all participants of the Zurich workshop and the ECPR workshop of 2018 in Nicosia, Cyprus. Reinout van der Veer thanks Markus Haverland and Michal Onderco for insightful ­comments, and the Netherlands Organisation for Scientific Research for financial support. Claire Dunlop and Claudio Radaelli acknowledge support from the project “Procedural Tools for Effective Governance” (ERC grant no. 694632) and extend particular thanks to Sébastien Chailleux, Cleo Davies, Eva Kunseler and Patrick Marier. Ignacio Sánchez-Cuenca wishes to thank Marina Costa Lobo, Robert Fisham and Adam Prezworski for comments. Silvana Târlea and Stefanie Bailer acknowledge funding from the EU’s Horizon 2020 research and innovation programme for the project EMU Choices (grant no. 649532) and thank Julia Dürr, Lara Eigenmann and Irina Fehr for research assistance. For comments on earlier versions of his chapter, Pier Domenico Tortola wishes to thank participants in the 2018 APSA (American Political Science Association) panel on “Concept Formation and Comparative Historical Analysis” and the 2018 ­colloquium in

xvi  Preface and acknowledgements European Politics and Society at the University of Groningen. Marina Costa Lobo and Ian McManus’s work was funded partly by Project MAPLE (ERC grant no. 682125). Despina Alexiadou wishes to thank Hanna Back and Patrick Dumont for their extensive comments, as well as Hakan Gunaydin for help with her chapter. Marco Valbruzzi thanks Stefano Bartolini, Duncan McDonnell and Gianfranco Pasquino for comments on early versions of his chapter. We are extremely thankful to Sonia Alonso Saenz de Oger and Sebastián Lavezzolo for their critical discussion at the book panel at the Council for European Studies in Madrid in 2019, as well as to Yves Mény for his insightful comments on the manuscript. For  their guidance through the publication process, we are grateful to the editors of the Routledge series on “Social and Political Elites”, Patrick Dumont and Keith Dowding, as well as to the editors at Routledge, Andrew Taylor and Sophie Iddamalgoda. Helen Belgian Cooper did a magnificent job copy-editing the manuscript. We are also very grateful to the anonymous reviewers of the draft manuscript for their insightful and constructive suggestions. Eri Bertsou and Daniele Caramani

Abbreviations

ANSES French Agency for Food, Environmental and Occupational Health and Safety BSE Britain Stronger in Europe BU Banking Union CEE Central/Eastern Europe CLEA Constituency-Level Data Archive CSR Country-specific recommendation EBA European Banking Authority EC European Commission ECB European Central Bank ECJ European Court of Justice ECM Error correction models EFSA European Food Safety Authority EFSF European Financial Stability Facility EIU Economist Intelligence Unit EMU European Monetary Union ERA European Regulatory Agency or Agencies ESFS European Financial Stability Facility ESM European Stability Mechanism EU European Union EVS European Values Survey GIIPS Greece, Ireland, Italy, Portugal and Spain IFI International financial institutions IMF International Monetary Fund INPS Italian National Social Pensions System ISI Import substitution industrialization

xviii  Abbreviations LDA LiTS MEFP NGO OECD OMT PR

Latent Direchlet Allocation Life in Transition Survey Memorandum of Economic and Financial Policies Non-governmental organization Organisation for Economic Co-operation and Development Outright monetary transactions Proportional representation

QE SGP SMaPP SSM UN VAT WTO WVS

Quantitative easing Stability and Growth Pact Social media and political participation Single Supervisory Mechanism United Nations Value-added tax World Trade Organization World Values Survey

Introduction The technocratic challenge to democracy Daniele Caramani

Introduction: technocracy in democracy’s shadow Modern democracy stems from the “twin” phenomena of mass mobilization that took place in the nineteenth century throughout the West. Together, the National and Industrial Revolutions characterize the most fundamental event of political and social mobilization of modern times.1 They represent the transition from local to national identity, from status to contract and from folk community to urban society. The  former brings about the nation-state as the fundamental unit of political organization, with democracy replacing autocracy. The latter establishes capitalism as the basic production mode, with industrial economy replacing agriculture. This double mobilization created the two most encompassing communities of our time: one cultural (the nation) and the other economic (class). While parallel, the two junctures bear within them a tension. The  National Revolution is at the origin of the fiction of the demos and their sovereignty. Nationalism as its main ideology is democratic, blurring the vertical stratification of society. The nation is fundamentally egalitarian. It is also intrinsically irrational.2 The Industrial Revolution carries forward another centuries-old fiction, a “utopia” really,3 of running society—primarily, but not only, the economy— like a machine, guided by the positivist belief in progress, knowledge and merit instead of caste privilege. Managing complexity is fundamentally non-egalitarian. It is also imbued with a belief in rationality, embodied by a state applying the procedures of industrial production modes through bureaucratic and territorial organization. The industrial confidence of humans controlling energy and, later, Taylorism spills over to managing scientifically social institutions.4 The  new reality emerging from this double transformation requires legitimacy, but also efficiency5; it is national as well as rational; it is, therefore, popular and elitist at the same time: a combination of nation and state. Representative government (rather than “democracy”) is the result of efforts to reconcile this duality—and it is more realistic with regard to the persistence of societal conflict, something both nationalism and scientific management do not contemplate. Initially, the intention to maintain elite autonomy vis-à-vis the

2  Daniele Caramani people is quite explicit. In very Burkean ways, self-rule institutional mechanisms restraining “trustee” representation are dismissed: direct democracy, the imperative mandate, the lot and other participatory democratic institutions. Instead, one finds indirect elections (for example, through electoral colleges) and restricted franchise based on wealth and education.6 With the extension of suffrage and with the introduction of executive responsibility towards parliament, Madisonian divisions of power, checks-and-balances and public areas deliberately isolated and “de-politicized” (such as independent judiciary, rigid constitutions, international law, state bureaucracy and central banks) are maintained to temper Rousseauian sovereignty. Most importantly, perhaps, the election as basic selection mechanism for the elite triumphs. The goal, once again, is to balance legitimate self-rule and efficiency, people and elite, popular will and an informed, competent, rational government.7 Electoral competition allows selecting the “best”, unlike the lot, and becomes the centrepiece of Schumpeter’s definition of democracy, with elites competing for the popular vote. A  competition for the popular vote, but elites nonetheless. Political parties, in particular mass parties in the twentieth century, come to play a central role in linking the enfranchised democratic masses (selfrule) to governmental action (efficiency) and securing responsiveness through the electoral sanction.8 Over the past 200 years there have been moments when the balance of this arrangement has been “challenged”. On the one hand, the very mechanisms that ensure minority protection, checks on the “tyranny” of the majority and the competitive selection of the elites are perceived as restrictions on the “will of the people”, and as fragmenting it into self-interested factions. On the other hand, the very same mechanisms are held responsible for the lack of efficiency of political competition, the short-sightedness of government (aimed at pleasing electorates in view of (re-)election), incompetence and, again, fragmentation in particularistic factions detrimental to the orderly functioning of society. The model according to which the balance is maintained is therefore under attack from the two sides it is meant to keep united: the popular and the elitist—i.e. the populist and the technocratic (Caramani 2017).9 While the most brutal challenge took place roughly 100 years ago during the build up towards fascist and communist totalitarian regimes in the 1920s and 1930s, representative democracy in the first two decades of the twentyfirst century is again being confronted. Academic work and the media have hammered on the populist challenge.10 In spite of an impressive literature on technocracy, the recent debate has neglected this additional challenge that has continuously side-lined representative democracy since its inception.11 The  goal of this book is to rebalance the debate. Recent literature has addressed the double challenge to representative democracy (Bickerton and Invernizzi Accetti 2017, Caramani 2017), yet the technocratic side is still under-researched in its own right. “Challenge”, therefore, means first and foremost a challenge to representative democracy. However, one should not forget the challenge that technocracy poses

Introduction: the technocratic challenge  3 in reaction to a more “radical” idea of democracy based on aggregative self-rule rather than the selection of competence. This is a challenge to “illiberal” populism expressed in one will and through one voice. In rebalancing the debate, one should not forget that both challenges—the technocratic and the populist—have much in common. While this is not a book on populism, technocracy, as we shall see, is the other side of the same medal.12 This  Introduction starts by defining “who” the challenger is—i.e. technocracy. It then addresses the question of “why” it has recently resurfaced. Next, it discusses “what” aspects of representative democracy technocracy challenges, as well as by which means—namely, depoliticization, holism, neutrality, responsibility, unaccountability, exclusion and autonomy. It  then considers “how” strong the challenge is, and “how” it manifests itself. Its multiple manifestations as regimes, institutions, actors, policies, discourses and individual attitudes are addressed. The conclusion asks whether technocracy is a threat or a corrective to the perceived shortcomings of representative democracy.

The challenger: technocratic power and representation If one defines politics as the conflict for the power to “allocate values” (i.e. decide on the “goals” of policies),13 then technocracy is a form of power in which decisions over the allocation of values are made by experts or technical elites based on their knowledge, independently and in the long-term interest of the whole of society.14 From this definition, a number of specific features follow. • •

• •

First, the source of legitimacy of power is superior knowledge and expertise (rather than, say, popular will). Second, political representation follows a “trustee” model positing independence from, and unresponsiveness towards, society, which is unable (because of incompetence and lack of time) to “mandate” decision makers and thus implement democracy as sovereign self-rule.15 “Democracy” is reduced (or “impoverished”, “devalued”) to a procedure of elite selection. Third, policy is geared towards acting in the general, common interest of the whole (and also long-term prospectively, in time), rather than its parts. Fourth, this interest is established rationally and scientifically (“objectively”), rather than derived from the aggregation of subjective, plural, shortterm interests.16

Technocracy is thus a type of power, but also a source of legitimacy and a form of representation. Were “technocracy” only to implement decisions on goals made by others (say, a democratic majority) and to remain under their control, then it would stop being technocracy—as no power, no goal-setting, and no allocation is involved, but only execution, implementation and advising on the means. It becomes then expertise, technicity, competence and knowledge. When skills are put to the benefit of the exercise of power, expertise becomes technocracy. Defined as a form of power, technocracy cannot be apolitical or neutral

4  Daniele Caramani (the fact that it is often presented like that is discussed later), which explains to a large extent the derogatory connotation the term assumes. Power in technocracy derives precisely from abandoning technical neutrality by setting the goals beyond the means.17 At its core, technocracy is elitist. It identifies an elite—based on its knowledge, expertise, superior academic credentials and experience—separate from ordinary people. The elite can best, and therefore should, guide society. It should act based on its judgement, disregarding the preferences of ordinary people who are less equipped in terms of skills and time or who are susceptible to ulterior motives. The knowledge elite is also better placed to provide effective solutions to complex social problems unconstrained by short-term partisan and electoral interests. Not  being bound (i.e. responsive) to the electoral approval of given social groups, it can act responsibly with a long-term view for the betterment of the entire community. This anti-plural scepticism towards what is seen as “particularistic” interests contrasting the general one implies that the elite does not  include the political class selected through parties and elections. In the technocratic view, there are decisions that are objectively “good” (or “bad”) for the whole (needless to say, they cannot be “left” or “right” as they would favour a part). Policies need not result from the aggregation of plural interests.18 Rather than being based on “inputs” from society, political action is established through rational speculation based on the belief of positivist “truths” to be established scientifically in the identification of both problems and solutions. As a form of power, legitimacy and representation, technocracy is therefore a multidimensional elitist, holistic and scientific phenomenon. As subsequent sections try to clarify, technocracy manifests itself in political systems to different degrees. Rarely is an absolute, maximalist conception in the form of a fully “technocratic state” observed or even championed.19 All regimes rely on some form of popular mobilization and inclusion. The  complete absence of technocratic power from political systems is also hard to imagine. Even the most “radical” democracies imaginable are likely to rely on some independent expertise. Technocratic power, rather, takes various grades, including the power of experts in advisory positions, independent ministers and prime ministers in cabinets, or entire executives. At a minimalist, “atomic” level, even a single actor can act technocratically or not, as well as make some technocratic statements but not others.20 This idea of technocratic power as a continuum then has important consequences for whether or not it is authoritarian. As technocracy is a quality of all regimes, it is only from a certain threshold that the level of technocratic power qualifies a regime as authoritarian.

The origins of the challenge The origins of technocratic power lie, as mentioned, in the necessity for modern states to operate in the changing and ever more complex society that issued from the Industrial Revolution. In  parallel to championing nationalism and

Introduction: the technocratic challenge  5 democratization, liberalism must also assure the efficient functioning of the “system” and its effectiveness in complex environments and on complex matters (Centeno 1994: 235). A double exchange takes place. On the one hand, the state learns to make use of the rational procedures and scientific methods from the advances in science and technology (most notably through data collection and the organization of bureaucracy and territory) and applies them to steering societies from industrial to educational and health policies.21 Political fragmentation, in Europe in particular, has contributed to the development of the technocratic capacity of states, given the harsh international competition (Tilly 1990).22 On the other hand, it is political fragmentation and international competition that drive technological progress and, ultimately, the Industrial Revolution itself.23 This exchange reinforces both state building and technological progress, and both feed one another. To the question, therefore, about the “why” of technocracy, the answer is complexity and competition. To the more specific question addressed in this volume, however, about the “why” of the challenge to representative democracy, the answer must be sought in the perceived incapacity of its main actors and institutions—parties and parliaments—to provide such expertise. The origins of the challenge stem from the changing nature of representative democracy, elitist critiques at the turn of both the twentieth and twenty-first centuries. Besides the function of representing by linking citizens, representatives and policies, political parties fulfil the function of governing responsibly and competently by forming political personnel, running governmental agencies and taking decisions. This assures output legitimacy. When parties cannot provide the necessary expertise, the technocratic challenge emerges. Yet until recently the critique of parties centred on their incapacity to represent, not  to govern. The transformation of parties from “mass” to “cartel” (Katz and Mair 1995) has been used to stress their governing functions, to the detriment of responsiveness to the electorate and policy congruence. In a growing literature, it is input legitimacy that has been questioned.24 It is therefore the populist critique of representative democracy that has been stressed. This grossly neglects the elitist critique of representative democracy, which has only recently received renewed attention (Caramani 2017, Habermas 2015). What this alternative critique stresses is the over-responsive nature of parties and the almost exclusive concern with being in constant “touch” with the electorate through increasingly sophisticated means to monitor public opinion, including polls, marketing tools and social media. Rather than having lost mobilization capacity, parties have acquired new means to persuade voters and foster loyalty through innovative channels of communication. According to this view, parties as “electoral machines” become less responsible because they act predominantly to seek popular consent. Once election is secured, policies are tailored to favour re-election. This leads to short-term responsiveness fed by the media’s hunger for immediate results in high-frequency electoral cycles. This critique is what causes calls for experts to take over. Technocracy is invoked as it involves decision makers who are not subject to the tyranny of popular consent.

6  Daniele Caramani Taking into account functionalist theories of technocracy (Williams 2006), this reaction to what is seen as pure “electoralism” is reinforced by two broad factors. First, the perception of the increasing complexity that supra-national and non-majoritarian governance involves, with intersecting cross-territorial jurisdictions, international organizations and courts, trans-governmental networks and agencies, and independent regulatory authorities.25 Such complexity calls for more technical competence, less public involvement and more managerial roles than decision-making ones (given the policy constraints that such multilevel and trans-territorial arrangements involve). A specific area of complexity is reforms needed either to make a country more competitive or to bring it in line with compliance norms required by supranational integration. Second, dissatisfaction with representative democracy is high when the results of policy action are poor. This is particularly true in times of crisis, be it in the aftermath of World War I or since the Great Recession in the EU that started in 2008.26 As Chapter 12 by Eduardo Dargent shows, it applies also of periods of crisis in Latin America, be it the appointment of “Chicago Boys” during the Chilean military regime (Silva 1991), the various technocratic appointments in Argentina, Brazil, Mexico and Peru in the 1980s and 1990s (Geddes 1990, 1994, Kaplan 2013, Teichman 1997, Weyland 2002) or today’s appointment of neoliberal economists in the new Brazilian cabinet. The demand for output legitimacy, effectiveness and efficiency is not only a reaction to parties and their role in representative democracy, but also to populism (the other parties’ great enemy) and its reliance on self-rule, its critique of elites and its scepticism towards expertise.

The objects of the challenge The arguments presented thus far to address the reasons for the challenge to representative democracy already point to “what” is challenged—namely, the basic features of representative democracy (Dahl 1956). By offering an alternative vision of power, legitimacy and representation, technocracy criticizes the submission of policy to democratic self-rule and competitive elections. This ­section addresses the features of representative democracy that technocracy directly challenges. The challenge to “politics”: depoliticization The  challenge of depoliticization is directed against the very essence of the political. Technocracy’s vision is the removal of “politics” from “policy”, and it points to a Hegelian vision of a universal state above and beyond clashes of interests. It sees itself as anti-political, external to, or “above” conflict, confrontation and opposition. First, there is a firm belief in the existence of policies that are “objectively” either good or bad for society, rather than different opinions/ preferences on what this could be. It is a matter of “facts”.27 Second, there is a

Introduction: the technocratic challenge  7 firm belief in the possibility to “discover” good policies and solutions rationally, based on analysis and scientific method, and using empirical data. This identification does not require politics as there are no differences possible in rationally established “truths”. Depoliticization, the non-ideological nature of technocratic representation and pragmatism all point to the superfluity of “choice”. Policy does not involve choice in the technocratic vision as the course of action needs simply to be identified—not chosen. Policies are therefore often presented in terms of necessity, pointing precisely to an impossibility to choose. While some authors (see, for example, Tortola in Chapter 3) define depoliticization in terms of detachment vis-à-vis conflicting contestants with different positions (that is, vis-à-vis “politics”), the challenge of depoliticization can be framed more radically, insofar as it does not contemplate conflict in the first place. There is nothing to be neutral about. Many of the movements that may be ranged under the broad label of technocracy have conflictless societies as their ultimate goal. This applies to Marxism, but also to the scientific management of society as dreamt of by proponents of technocratic soviets or committees on technocracy in the United States (Thorstein Veblen and Walter Rautenstrauch among others). It continues with “problem solving” and “epistocracy” approaches aiming at “neutralizing” conflict (Bersch 2016, O’Donnell 1994)—in the sense of abolishing conflict, rather than taking a neutral position within a conflict. The political process is equated to a matter of technique; not conflict, aggregation and compromise. The very idea of non-majoritarian institutions and the “regulatory state” making rules and monitoring and implementing them (Majone 1994), is one that does not necessitate political decisions but rather planning and dirigisme.28 Even the selection of personnel is not conflictual as it is based on objective meritocratic standards. Finally, constitutionalism—the institutionalization of welfare and market—can be interpreted as non-conflictual solutions to the allocation of resources.29 The challenge to pluralism: holism The point that technocracy identifies solutions in a non-conflictual way implies that the “whole” takes precedence over the “part”. What is good or bad is meant to apply to society in its entirety, not to a segment of it, and does not contemplate “trade-offs” or winners and losers. Conflict, then, arises only when a part is directed against the interest of the whole, not against another part. The technocratic vision is about the common interest embodied by the state, and its political decisions are meant to be decisions for all rather than allocation, or aggregation, between diverse preferences. To be sure, an aggregation takes place, but according to rational rules similar to those of a machine working through its different components, not competition. What rational speculation and scientific procedures achieve is the identification of a general interest and the measures to fulfil it. Technocracy, as populism, is therefore anti-pluralist in the sense that it does not allow the interest of the part to prevail over the interest of the whole. As a consequence it is anti-partisan,

8  Daniele Caramani as parties are carriers of particular interests rather than of the interest of society in general.30 The “mediation” of groups’ preferences is seen as distortive of the common good. To be against a rationally defined policy is to be against what is “right”.31 Opposition is thus illegitimate on the grounds that it is irrational. There is no place for political cleavages, only for socioeconomic diversity. The challenge to ideology: neutrality Political ideologies are, on the one hand, interpretations of the social world and, on the other, projects for change and political action. Political parties with governing ambitions propose alternative visions for the aggregation of the plurality of interests. For  such parties there is a holistic vision of society.32 What distinguishes representative democracy from technocracy (and populism) is that there are competing proposals on what is good for society. Ideologies are packages of policy alternatives, so voters can make a meaningful choice. Importantly, voting binds the party to the electorate in a quasi-contractual or mandate relationship. It implies a commitment to a mandate based on “promissory” proposals (Converse and Pierce 1986, Mansbridge 2003, Pierce 1999).33 Ideas of electoral mandate and choice do not apply to the technocratic view of power. First, the electorate does not have the competence to mandate policies, or even to choose between them. Second, it makes no sense to give a choice if through rational speculation it is possible to identify “the” best one. Technocracy therefore has the pretension of being a-ideological. Its discourse is one of “pragmatism” and neutrality (Adolph 2013).34 Yet even if this discourse is presented in non-ideological terms, as Sánchez-Cuenca shows in Chapter 2, technocracy is in practice never neutral. All production and transmission of knowledge is a socially constructed activity that reflects values and power (Feindt and Oels 2005, Fischer 1990, Foucault 1980, Jasanoff 1990). Three points should be made here. First, historically, one can think of socialist planned economies, fascist corporatist states and neoliberal institutions as examples of technocracies. The 1960s and 1970s were characterized by “developmentalist” experts (on agriculture, health, social policies and poverty alleviation)—especially in Latin America, as Dargent shows in Chapter 12—but were replaced by economically “neoliberal”, “orthodox” economists in the 1990s (Babb 2001, Centeno and Maxfield 1992, Grindle 1977, Sikkink 1991). Technocracy, in this sense, can therefore be conceived of in terms of a “thin ideology” that can be filled with different content, similarly to populism. Second, more problematic are cases of policies presented as necessity, when political action is guided by external constraints that hinder the possibility of alternative policies and a real choice. Both access to international markets, credit and trade, on the one hand (the politics of “adjustment” as labelled by Kahler 1990, 1992), and supranational integration and compliance, on the other, reduce the latitude of possible economic and social policies. This drives them towards liberalization, privatization, finance reforms and “austerity” guided by international bodies such as the European Commission (EC), the International Monetary

Introduction: the technocratic challenge  9 Fund (IMF) and the World Bank or required by membership in the World Trade Organization (WTO). Such policies go under the “neutral” language of “reforms” and “modernization” in spite of hiding a neoliberal agenda (Crouch 2011).35 Third, as instrumental theories of technocracy argue, technocrats may be agents of business and international financial institutions, or even of politicians who use them as “tools” to carry out reforms (Magaloni 2006). This means that they may not be autonomous, a challenge discussed later in this Introduction. The challenge to representation: responsibility, unaccountability, exclusion Technocracy also challenges the basic principles of representation—namely, that decision makers should be responsive to the electorate, accountable at regular intervals and inclusive of all segments of society. Instead, from the technocratic perspective responsiveness must make way for responsible governing, voters do not have the competences to sanction elites, and representatives should include only the “best” selected through merit. 1. Responsibility. The literature has recently stressed the focus of technocracy on responsibility and governing, if necessary, disregarding responsiveness to any sort of “mandate” from the electorate and incongruent with its preferences (Caramani 2017). Proximity with the distribution of the electorates’ preferences should not guide policy as the people do not possess the faculty of identifying society’s general interest. This “trustee” vision of representation is clearly inimical to populism, but also to the self-rule component of representative democracy in which parties compete based on proposals. While party government is mainly based on a prospective “mandate”, technocracy is not. Even more so, this view is damaging to direct democracy. Complementing the literature’s focus on the unresponsiveness of technocracy, the arguments presented by Van der Veer in Chapter 4 lead to a more nuanced picture.36 First, the usual acceptation of responsiveness is too much linked to congruence with the majority’s preferences, neglecting responsiveness to minorities as well as domestic and international markets asking for stability. Second, elites need to respond not only to preferences but also to risks—in particular societal, institutional and reputational risks.37 Third, technocrats’ actions are not necessarily unresponsive to citizens when they do not follow a mandate received democratically. In contexts of high levels of patronage and corruption, or of weak bureaucracy and low stability, elected politicians are often even less responsive to citizens’ demands than technocrats (Dargent 2015). In such contexts, public preferences are disregarded in favour of advancing particular interests, thereby losing sight of the general interest, as would be the case for structured ideological parties. 2. Unaccountability. The same arguments extend to the unaccountable nature of technocracy. First, since it does not  govern based on a programme

10  Daniele Caramani (mandate), it is not possible to ascertain whether or not the promise has been maintained. Second, if technocracy operates through superior knowledge and expertise, it makes no sense to ask less competent voters to judge policies established by rational procedures. Technocracy is thus unaccountable.38 While a crucial feature of representative government, the technocratic critique is precisely that parties and leaders operate in “anticipation” of the electoral sanction—therefore once again tailoring their policies to please rather than to solve.39 Again, one needs to nuance such arguments, as, while perhaps not capable of judging technicalities, voters are still able to assess whether or not problems have been identified and solved. A form of vertical accountability (the  possibility for voters to sanction elites) is therefore plausible. If  legitimacy is based on output, then an evaluation of results can be expected, if not of the means to reach them. Furthermore, technocratic elites are not  monolithic. Other experts (academics or journalists) can air their critique on measures and results. This would point to the possibility of some form of horizontal accountability in which elites are constrained by forms of quality control, accuracy of scientific protocols, “peer review”, data sources and fact checking. 3. Exclusion. Finally, technocracy challenges our understanding of representation in its descriptive dimension—this, however, up to a certain extent only, as representative democracy is not primarily intended as a “mirror” of the diversity of society or a “mini-society”. The focus of representative democracy is on “doing” and “acting” rather than on “being” and “standing for”.40 The focus on elections, rather than the lot, to achieve a superior quality of representation in terms of “aristocratic democracy” (Manin 1997), and the “politics of ideas” rather than “of presence”, all point to a scepticism towards the inclusion of groups of a descriptive basis in representative functions.41 In technocracy, this scepticism is pushed to the extreme but does not represent a qualitative break with representative democracy. Descriptive representation should be minimal, and technocrats should be selected through proof of merit (selection of the “best” and “talent”), rather than mirror society’s sociological diversity. Representation is based on competence rather than programmes, and therefore trust is based on being different—in education, intellect and expertise—rather than on similarity (as would be the case in populism). The  principle is one of exclusion of the masses, rather than their inclusion through self-rule, and of isolation of decision makers from the masses’ electoral process.42 The challenge to self-rule: autonomy (independence) Technocratic power needs to be autonomous otherwise it is not a form of power but simply an agency for implementation. In this, technocracy seems to pose a challenge more to democracy as self-rule, i.e. more to populism than to representative

Introduction: the technocratic challenge  11 democracy that combines the conception of democracy as self-rule with that of selection of representatives (see Chapters 1 and 2). There is a further aspect that technocracy and representative democracy share, in opposition to populism— namely, the recognition of autonomous (independent) spaces of decision making beyond popular control. Autonomies are enshrined in fundamental rights beyond the reach of electoral majorities or direct democracy, but can be found also in the autonomy of central banks, the judiciary, state broadcasters, certain ministries and bureaucracies, and diplomacy. This challenge is therefore directed primarily towards one part only of representative democracy, with which it shares the feature of autonomy, that is towards a more radical idea of “democracy”, the populist idea of democracy as sovereign self-rule. Of course, technocracy as regime implies a far-reaching autonomy for nondemocratic decision making, whereas under representative government such autonomy is more limited. It seems, nevertheless, to be a matter of degree rather than a qualitative break. In  either case, the challenge that autonomy poses to democratic government is twofold—on the one hand, when technocrats are considered as actors, or, on the other, when they are considered as agents. First, on a general and abstract level, technocratic autonomy challenges both representative and “radical” versions of democracy insofar as it isolates areas of decisionmaking from democratic control. The  broader and more central these areas, the stronger the “democratic deficit”. The goal of autonomy is to “shield” decisionmaking from partisan influences, and central banks’ independence has the express goal of avoiding politicians “pandering to the people” by using the monetary printing press for popular consent (see Chapter  4). The  necessary consequence is that it undermines the democratic legitimacy of such decisions. This  is the challenge coming from technocrats as “actors”.43 The  asymmetrical expertise between politicians and citizens on the one hand, and experts on the other, gives the latter great power. The fact that politicians depend on their expertise (think of the military) gives technocrats an advantage in strategically advancing their own interests and acquiring even more autonomy in pursuing goals that are not reflective of the demands and interests of society, even if they are objective and neutral.44 Second, on a more specific and concrete level, formal autonomy does not necessarily mean that technocratic bodies are also autonomous in practice. Technocrats may not need electoral approval but, nonetheless, they need to be appointed by parliament and its majorities. Rather than autonomous, such bodies and personnel become “agents” of specific interests. Autonomy from democratic control does not mean autonomy from other “principals” such as politicians who use them to carry out the “dirty job” of difficult reforms leading to “vales of tears” (Di Palma 1990) or lobbies. Indeed, instrumentalist theories maintain that technocrats are the executioners and representatives of dominant groups in society (Centeno and Silva 1998).45 The  challenge to both representative (its selfrule component) and “radical” versions of democracy is that a pretend autonomy empowers corporations and private interest groups.

12  Daniele Caramani

The strength of the challenge How strong is the challenge, and what are its manifestations? Technocracy manifests itself in political systems to different degrees. Empirical instances of technocracy range along a continuum. Chapter 1 by Bickerton and Invernizzi Accetti (following Radaelli 1999a) opposes a classical undemocratic argument for technocracy and a contemporary argument respectful of democratic values and institutions. Furthermore, manifestations vary between explicit institutions (either the transformation of existing democratic ones, such as cabinets, or new bodies, nationally and supra-nationally) and more diffuse ones as part of a discourse. At one extreme of this continuum, in a fully “technocratic state” only experts set the goals. At the other, there is no role whatsoever for independent expertise, and no goals are set other than by popular will, with experts involved only in implementation. Empirical cases fall between these extremes. The complete absence of technocratic power is hard to imagine. Every democracy has technocratic pockets in the form of state bureaucracy, experts in advisory positions, independent ministers and prime ministers or entire executives. Conversely, even the closest cases of technocracy (say, soviets or theocracies) must rely on some degree of popular mobilization. Technocratic power as a continuum has important consequences for its authoritarian nature. The  technocratic challenge may remain within the democratic system or, on the contrary, may transcend it, thus becoming “anti-system”. It is only from a certain threshold that the level of technocratic power qualifies a regime as authoritarian. How much technocracy a specific political system entails is therefore an empirical question. Also, the support technocratic ideas receive is a matter of empirical investigation. In both cases, the system’s technocratic nature, and the support for it, must be addressed at different levels: the state as a whole, its institutions, the personnel (“technocrats” and their role in policymaking), their discourse, and citizens’ support for technocracy.46 These levels are discussed next, and build the framework of the empirical analyses in the volume. The technocratic state Technocracy needs a state. It operates through a state organization that provides the apparatus to decide and to act efficiently and effectively. •

•

First, the state’s cybernetic function is to “think”—i.e. to acquire information about society and the environment, process complexity and organize rational procedures to define goals and means to establish good, long-term, responsible policies (see also footnote 21). Second, it is states that act in a constraining international system in which “adjustment” to market forces (to secure credit, as dependency theory would stress) and compliance to agreements within international financial institutions (IFI) are required; this is especially relevant for modernizing countries in the world’s periphery, but also in supranational negotiations (as Chapter 8 by Târlea and Bailer discusses for EU’s member states).

Introduction: the technocratic challenge  13 •

Third, internally, technocracy needs the state (repressive) apparatus to impose its will and implement decisions, in principle against the will of the people when necessary.47

Besides being efficient and effective, strong state structures assure experts’ independence from society’s dominant forces. As Centeno (1994: 34) shows, a technocratic state cannot afford to be isolated (as it needs information about the society it acts upon) but it must be insulated (i.e. an autonomous “actor”). Very much in line with neo-institutionalism, the technocratic state is not the reflection of dominant interests, or an instrument or “agent” of particular social groups. It must, nevertheless, be a state that has a “network” relationship with different parts of society, the different components of the “body”, in a corporatist way, and be receptive to their needs. The linkage between state and society is often assured by a single mass party. This  has been the case in a number of examples that would qualify as technocratic states—namely, modernizing countries such as Mexico, China, South Korea, Singapore, Taiwan and the Soviet Union (Hoffman and Laird 1985), making it an electoral-authoritarian form of governance. Certainly, a maximalist definition of technocracy in terms of state makes it an authoritarian challenge to democracy. Yet not  all forms of authoritarianism are technocratic. That a technocratic state cannot be purely repressive—insofar as it needs a linkage to society’s plurality of interests—would lead to the exclusion of purely military regimes from this type of state. High state capacity and insulation does not necessarily mean that the state is more technocratic. Historically, military regimes have had the need to rely on technocracy. Yet the Chilean dictatorship was more an experiment in implementing neoliberal policies, which required limiting democratic self-government, than a technocracy. As Sánchez-Cuenca stresses in Chapter 2, neoliberalism presupposes a state that frees markets from democratic interferences. This, however, does not make the state technocratic—on the contrary, it imposes the rule of the market. The  classical literature on technocracy has repeatedly stressed its ambiguous relationship with authoritarian military governments or semi-authoritarian bureaucratic regimes (Collier 1979, O’Donnell 1973), especially in regions such as Latin America. Yet the challenge to democratic institutions persists also during periods of democratization, such as the 1990s, when scholars noted that the influence of technocrats persists also in democratic settings (Centeno and Silva 1998, Domínguez 1997). Experts remain relevant actors under democracies, and even stabilize their influence (Dargent 2015).48 A  number of authors stress the limiting effect of technocratic influence on democratic politics, and speak of “technocratic democracies”.49 Policy remains in the hands of technocrats, often against campaign promises and programmes. This leads us to conclude that the line between an authoritarian technocratic state and technocracy under democratic institutions is a blurred one. Besides utopian nonsense, there can only be arbitrary thresholds from whence on a state qualifies as technocratic.

14  Daniele Caramani The democratic state and technocracy Autonomy does not necessarily mean that states become authoritarian. Autonomy exists also in democratic states, to a greater or lesser extent. The  ministerial bureaucracy is a case in point, but also central banks and various regulatory agencies. Expertise is present too in the legislature, with various offices supporting parliamentary works through the preparation of dossiers and the briefing of parliamentarians. Parties themselves provide expertise. Under the broad principle of the separation of power, the independence of the judiciary and its increasing policy impact—i.e. the “judicialization” of politics—is another area of autonomous state activity.50 What follows in this volume does not address all these vast areas within a democratic state that are shielded from political contestation and electoral legitimation as they do not pose a challenge to representative, procedural democracy. This volume, however, does address the challenge of extending areas of technocratic autonomy to the executive (Chapter 6), and the ensuing policy consequences (Chapters 7 and 8). This is particularly important as it is the branch in which policy is enacted (as well as, to a large extent, initiated). Recent decades have seen the rise of technocratic or technocrat-led cabinets as opposed to the more or less ideal-type of party government (Katz 1986). Cabinets of this type have formed in a number of countries, including Bulgaria, the Czech Republic, Finland, Greece, Italy, Lithuania, Portugal and Romania in the 2000s and 2010s, following economic crises, scandals, failure to form coalitions, etc.—in general, as a consequence of “party failure”.51 Once more, the picture must be one of degree, in which the “non-partyness” of a cabinet runs from a maximum to a minimum (Andeweg 2000). Recent literature has specified criteria to assess the more or less technocratic nature of executives (Brunclík 2016, McDonnell and Valbruzzi 2014, Costa Pinto et  al. 2017), and Chapter 6 by Valbruzzi further advances this task. Technocratic government takes place when governmental decisions are made by unelected officials, who are not recruited through parties, and when policies are not decided within parties. Empirical cases of cabinets then (1) range between being more or less technocratic and (2) cluster in types of technocratic cabinets, based on their composition, their output (the remit) and their duration (term of office). Typologies thus distinguish between technocratic cabinets that have a mandate to maintain the status quo (short-lived caretaker cabinets) and those that change the status quo and make policy. Here we find cabinets that are fully technocratic (i.e. the majority of the cabinet is composed of non-elected, non-partisan officials, including the prime minister) and technocrat-led cabinets (in which a majority of partisan ministers is led by an unelected one). Only these qualify as technocratic as they have the power to allocate values, unlike caretaker cabinets. Technocratic cabinets are one of the most visible expressions of technocracy. They are a manifestation of the anti-party nature of technocracy: the distrust of parties for their short-term, vote-seeking and particularistic nature. In addition, they are a manifestation of the critique of inefficiency of party government, with

Introduction: the technocratic challenge  15 its lengthy processes. This challenge remains within the democratic state as— unlike a technocratic state—technocratic cabinets are temporary and step in only when parties fail. Elections remain available to hold such cabinets accountable, as well as the parties that support them in parliament. The democratic linkage between voters and executive (especially in parliamentary democracies) is maintained. In fact, it is difficult in many cases to speak of a “challenge” as it is parties themselves that give up their governing function rather than being challenged by pro-active technocrats. Technocrats, technocratic bodies and technocratic policies One of the main issues in determining the composition of cabinets—in turn to establish their “technocraticness”—is to decide who qualifies as a technocrat in the first place. This issue, however, concerns not only cabinets, but state bodies in general. This  leads to a further manifestation of technocracy—namely, personnel in the form of “civilian” unelected officials, who are not professional politicians, have no political experience and do not belong to party organizations. Such a perspective presupposes a definition of who—individually or as a group of individuals—can be characterized as a technocrat.52 As a first element, the literature stresses a high level of expertise derived from specific training, education or professional experience (in business sectors, academia or elsewhere). Crucially, they are not experts in politics, but in some other area (Centeno and Silva 1998).53 This expertise applies to both “outsiders” (i.e. previously not active in state positions) and “insiders” (i.e. from state bureaucracy) called to occupy executive positions (Bersch 2016). While expertise is a necessary condition, it is not sufficient. As a second element, such individuals must occupy relevant state positions that are not exclusively ones of implementation and execution. Technocrats are more than experts in the sense that they have power over policy—i.e. in determining a direction of adoption or of veto. A  technocrat is an expert active in politics. According to Meynaud (1969), when becoming a technocrat, the expert becomes political and ceases to be merely an expert. A técnico without power is not, therefore, a tecnócrata.54 As a third element, crucially, such individuals are not  elected; rather, they are appointed (selected), and therefore they are not subject to the constraints of electoral competition and partisanship (Malbin 1980, Tucker 2018, Vibert 2007). Technocrats, to be defined as such, must thus lack experience of public office, party activity and membership, and identification with clear political preferences (for example, through media statements). Defining technocrats as experts with executive roles who see themselves as neutral and independent of political competition should not imply that expertise is absent in political actors, be it individual politicians or political parties. In fact, political parties have long played a crucial role in forming political personnel and have served as “schools” for the preparation of politicians. Among the “functions” attributed to political parties, besides the crucial ones of aggregation,

16  Daniele Caramani structuring of the vote and governing, there is also that of “recruitment” of competent personnel in various policy fields.55 What exactly are technocrats experts in? One important distinction in this regard is between the supranational and the national levels. At the supranational level, the technocratic nature of international organizations such as the EU, the World Bank and the IMF, as well as the United Nation (UN) agencies in general, appears more clearly. Their non-elective nature and technical expertise make them ideal-case institutions dominated by technocratic personnel. The  extent to which they remain autonomous from political motives (mostly coming from domestic influences) and from ideological bias is a matter of empirical investigation. In addition, there are examples from a variety of areas in which expertise is activated. Most policy areas are covered.56 This is not the case at the national level, where technocrats are primarily synonymous with economists (either from academia, central banks, private firms or international organizations). This  focus on economics and finance, obviously, needs not be. As shown in Chapter 12, the areas in which technocrats are appointed can vary over time and extend into health, social policy, education and poverty alleviation among others. It also varies regionally, with Eastern Europe experiencing more technocrats in the fields of EU integration, and Latin America in social policy areas. Economics is, however, the crucial field. Variation in the levels of “technocraticness” and, therefore, in the strength of the challenge to representative democracy also occurs at the level of policies and the policy process, as Chapter 10 by Dunlop and Radaelli makes clear. Chapter 7 shows that technocratic cabinets behave differently to political ones. More or less technocratic policy processes (in their utilization of knowledge and involvement of experts in public policy) take place across countries, but also within countries. There is variation between periods of time but also during the same period of time across policy fields. At  this meso level, it is possible to observe the more or less prominent incidence of technocratic actors such as scientific communities, technical bureaucrats, research institutes and think tanks— the epistemic community at large. The contribution of Chapter 10 is precisely to address the policy process, rather than institutions and personnel. Technocratic discourse Manifestations of technocratic discourse and of technocratic actions are not  exclusive to technocrats and technocratic bodies (such as cabinets). Nontechnocratic leaders and organizations (say, parties) can and often do hold a technocratic discourse, make technocratic statements and act technocratically.57 Political leaders often invoke their expertise and competence; cite experts, statistics, facts and reports; and refer to procedures, as well as to a general long-term responsibility. Programmes and policies are dubbed as irresponsible or shortsighted in a derogatory manner. This means that technocracy does not manifest itself exclusively through technocratic actors—non-technocratic actors can also have discourses that are technocratic.

Introduction: the technocratic challenge  17 As Chapter 5 on the levels of analysis makes clear, this is a micro level. At a minimalist, “atomic” level, even one single actor can act technocratically (or not), as well as make some technocratic statements (but not always). The basic unit of analysis becomes the single statement and the more general discourse that statements (and actions) form. References to expertise, competence, long-term responsibility, general interest, neutrality and independence from political contestation, etc., become a discursive “tool” available to political actors (whether or not they are considered technocratic), in the same way that non-populist actors can make populist statements. Indeed, the concept of discourse and discourse analysis has been developed a great deal in the study of populism, although under different acceptations (Poblete 2015). Post-structuralism in the tradition of Ernesto Laclau is the broadest and includes all social practices creating meaning, not  only discourse in a narrow sense. Discourse is broader than a semantic phenomenon and needs interpretation.58 Post-modernism, too, considers discourse to have broad and latent meaning, but positively identifiable in speech and writing although not  quantitatively (hence its reliance on holistic grading; see Hawkins 2009). Content analysis, finally, understands discourse as semantic, linguistic allocution, communicative performance and style (including complexity and erudition); ultimately, statements, sentences and single words in texts. Its manifestation is quantifiable either through manual coding based on a codebook or through automated software based on a dictionary.59 As a discourse, technocracy can be more or less explicit and more or less hegemonic within a given society. It can reveal a society project and national identity, and include symbolic elements.60 For the sake of identifying empirical manifestations (to be counted or interpreted), technocratic discourse can be broken down into dimensions: • • • • • •

Elitism includes references to expertise, the exclusionary nature of the elite and merit. The scientific approach to policy leads to statements about objectivity, rational speculation and “truth” (evidence), pragmatism and necessity (absence of choice). Policies are claimed to be for the whole society (rather than its plural parts) in the long term, to bring about progress for societal welfare and society as a complex of functions forming a body or machine. Goal orientation points to efficiency, cost-optimization, pragmatism and flexibility, focused on fixing problems. Anti-political discourse mentions the advantages of the non-elected nature of officials, therefore independent, non-partisan, above politics and non-divisive. Finally, technocratic discourse is associated with a certain style that separates the elite from the masses, including attire, technical jargon, and reference to data, facts, figures, statistics and deprived of emotional, dramatic and conflictual tones.

18  Daniele Caramani Chapter 5 presents an example of a codebook. Chapter 9 by Nava, Centeno and Liu juxtaposes the technocratic framing over Brexit to a “direct democratic” one based on the use of the referendum instrument. Such approaches make it possible to analyse technocracy in a much more diffuse way in societies.61 Also, it allows analysis of such discourse in populist actors to identify areas of overlap between the two, as well as contrasts in their discourse. Among the “actors” that may be characterized by such discourse, one can include single citizens. Technocratic citizens? Technocracy in single citizens, however, does not manifest itself through statements, but in their individual attitudes. As Chapters  11  and 13  show, citizens may have more or less favourable attitudes towards technocratic governance (and technocratic cabinets more specifically), which diminishes the support for liberal representative institutions, thus constituting a further, and more diffuse, source of challenge. Unlike citizens’ populist attitudes, centred on the belief that politics should be guided by the will of the people, unconstrained by procedures and elites,62 technocratic attitudes reveal preferences for delegating decisions to experts, if necessary disregarding the people’s will. Chapter 5 discusses specific examples of surveys tapping into such attitudes. Experts are perceived as having greater competence to address complexity in a responsible way, not misguided by short-term electoral interests, such as politicians needing to be responsive to uniformed citizens. Similar to populist ones, however, technocratic attitudes include scepticism towards party politics and representative democracy in general. “Technocratic citizens” (if one may give them that label) are elitist (rather than anti-elitist), believe in expertise as a guide to policy (rather than in the people’s will) and are anti-political. To date, there are few explorations of technocratic attitudes. The  first study using survey data was done on the technocratic mentality of bureaucrats (Putnam 1977), but its items on objectivity, political neutrality and the scientific view of policymaking can be used for citizens as well. The concept of “stealth democracy” (Hibbing and Theiss-Morse 2002) measures citizens’ preferences for less popular involvement and more effective decision making carried out by unelected experts or business people. The authors found that a large chunk of Americans welcomed a more detached and efficient way of governing that can bypass disagreement and division of ordinary politics.63 Recently, Bertsou and Pastorella (2017) used a survey item asking respondents if they consider having “experts, not  government” making political decisions to be a good form of governance, as a proxy for technocratic attitudes to investigate their determinants and sources of variation across Europe. Bertsou and Caramani (2017, forthcoming), administered a novel battery of items on technocratic attitudes in nine European countries. Very little is yet known of citizens who have positive attitudes towards technocracy, but it is possible to hypothesize that “technocratic citizens” have specific socioeconomic traits (such as higher education levels) and attitudinal traits

Introduction: the technocratic challenge  19 (such as lower levels of trust towards political parties and parliamentarians, coupled with high levels of interest and competence in politics). Indeed, initial results from the new survey point in this direction. The inclusion of variables such as left–right placement also allows testing of the correctness of the claimed “neutrality” of technocracy at the individual level, and thus predictions of voting behaviour of “technocratic citizens”. The larger the pool of such citizens in a democratic polity, the stronger the challenge to representative democracy as this would lack the kind of diffuse support necessary for citizens to participate in it and accept its legitimacy. Among all the challenges, citizens believing that they themselves should retreat from politics is certainly the most worrisome for a liberal conception of democracy, especially if coupled with a second group of citizens with populist attitudes believing the opposite—namely, that elites are necessarily bad and corrupt. Thus far only the populist challenge has been addressed in research; the addition of the technocratic one may lead to a bleaker picture yet for the risks that representative democracy, as we have seen it develop over the past two centuries, faces.64

Conclusion: the technocratic challenge as threat or corrective Over the past 200 years, modern politics has attempted to reconcile demands for popular participation intrinsic to the idea of a sovereign nation that rules itself (a demos) with demands for selecting leaders based on merit and for effectiveness intrinsic to the complexities of societies and an ever-fiercer competition between modernizing states. While democratic inclusion is a response to the former, the rational state is a response to the latter. Representative democracy is the synthesis of these two sets of demands. The  two-faced nature of democracy has been described in many ways, the most notable example probably being the “redemptive” vs the “pragmatic” faces of democracy distinguished by Canovan (1999) or the aggregative and the liberal, as Sánchez-Cuenca calls them in Chapter 2. The former is based on faith and salvation, the latter on rules and institutions. To take Dahl’s dimensions of democratization, it espouses broad inclusion (mobilization and participation) and competition (Dahl 1971); illiberal democracy the first, undemocratic liberalism the second (Mounk 2018). Legitimate and responsive, but also efficient and responsible (Mair 2009, Scharpf 1999); antagonistic democracy vs management and administration in a conflictless society (in Mouffe, Laclau and Rosanvallon). This duality—in terms of nation and state, populism and technocracy (Caramani 2017), or “sovereign” and “government” as Bickerton and Invernizzi Accetti, following Rousseau, conceptualize it in Chapter 1—was the starting point of this Introduction. Yet whereas populism has often been presented in terms of a shadow of democracy, or as a haunting spectre visiting from time to time (Arditi 2004), not so technocracy—and this volume attempts to complete the picture. Figure I.1 illustrates the “bridging” achieved by representative institutions and its most successful interpreters, namely “responsible parties” (McCall Rosenbluth and Shapiro 2018). Representative democracy shares with populism, although

20  Daniele Caramani REPRESENTATIVE DEMOCRACY

Reason

Will

POPULISM

Holism

TECHNOCRACY

Figure I.1  The  triangular relationship between representative democracy, technocracy and populism

with no perfect overlap, a broad inclusiveness and popular participation through various democratic channels (from mass parties and elections to direct democracy, plus other forms secured by civil rights, varying across space and over time). Representative democracy shares with technocracy, but again with no perfect overlap, effective governance and mutual control mechanisms in the form of checks-and-balances, separation of power, minority protection and areas of autonomy (independent judiciary and monetary policy). The largest distance is with the idea that society is “one”—a holistic idea that populism and technocracy share in their opposition to pluralism, partisan politics and particularistic interests. On the contrary, the largest distance for populism is from “reason”, or “truth” (Shils 1956), and the largest distance for technocracy is from the “will of the people”, i.e. the redemptive or aggregative conception of democracy. Different polities, at different moments, have oscillated between the poles of populism and technocracy. Even within representative democracy, there is variation between countries, as well as over time, along the continuum towards one or the other end. Beyond the representative democratic polity, no “radical” democratic community has ever obliterated completely some form of bureaucratic institutions that are autonomous from popular support or conflictual politics. Similarly, no polity—even those that came close to something of a “technocratic state”—could ever afford to avoid having some form of popular mobilization behind its structures. It would, however, be wrong to remain with a static image. Even during periods of stable institutional arrangements, demands towards either of the two extremes are constantly voiced. Current times are an instance of such voicing for more popular participation and, at the same time, for more

Introduction: the technocratic challenge  21 effectiveness—necessitating less popular participation. While representative democracy has long been successful in bridging these contradictory demands, its capabilities in doing so are put into question today. This is precisely the “challenge” this book addresses. First, to what extent is there a “challenge”, and when does it become a threat? Second, can existing representative institutions respond to these contradictory demands? A great deal of literature is devoted to the challenge of populism, i.e. a reaction to an asymmetry between pragmatic and redemptive sides when there is an excess of pragmatism. This book is a step towards a more balanced analysis including the other challenge or, possibly, threat—i.e. unresponsive, anti-pluralist and autonomous policymaking beyond democratic accountability, yet still influenced by dominant sectors of society with non-majoritarian ways to voice their demands. Presenting representative democracy as bridging demands for popular participation on the one hand, and rational and effective policymaking on the other, opens up a third question: when do demands for more technocracy cease being a threat and become a corrective? Much, again, has been written about populism as a corrective by reintroducing both contestation and participation (or inclusion).65 The same should be done for technocracy, and it is the task of the Conclusion to answer whether it is a friend or a foe to democracy. Against electoralism, overresponsiveness and the need to please the masses, as well as populism itself, calls for technocracy may have the “healthy” effect of injecting expertise, competence and long-term planning in an ever more complex environment. The corrective may act also in limiting majoritarianism against the protection of minorities. This is certainly the neoliberal vision based on the protection of freedom as the ultimate goal of democratic institutions (see Chapter 2). Procedures, rule of law, rights and institutions may need to find a renewed role in contexts of predominantly protest-like expressions of voice. Furthermore, depoliticization may be a necessary limitation when politics is so polarized as to make majoritarianism (especially in contexts of ascriptive cleavages) dangerous.66 Also as a corrective, technocracy can operate at different levels: institutions, personnel and more informed, competent citizens. Populism is a threat to liberal representative democracy when it stops making legitimate claims for more inclusion within the system, participation and equality, and the abolition of rules and rights that limit power and scupper fair competition. Technocracy is a threat when it stops being technical expertise, a source of information for citizens, media and parties to have a responsible debate, and trade-off evaluations about different proposals of “aggregation”, to replace it with an unaccountable chimera of scientific consensus.67 It  is a threat when it sets goals beyond democratic control—i.e. when it becomes fully autonomous. Representative democracy is an “experiment” by complex mass societies in balancing popular inclusion, participation, and responsive and accountable policy, on the one hand, with expertise, decision-making efficiency and policy effectiveness on the other—in other words, input legitimacy with output. Like the proverbial bicycle, its balance is a matter of continuous corrections. The technocratic challenge described in this volume can be considered either

22  Daniele Caramani as a counter-steering against the imbalance caused by the surge of populism, or “over-correcting” against representative democracy itself, in which case it becomes a threat to it. What seems certain is that a bicycle needs to move forward so as not to fall down. For that to happen, involved, competent and supportive citizens are needed, who identify with its institutions and feel ownership towards them.

Notes 1 The labelling of the two “revolutions” is taken from Rokkan (1970). The expressions “twin revolutions” is from Bendix (1964), Mann (1993) and McNeill (1963). 2 On the “invention” of the people, see Kalyvas (2005) and Morgan (1988), and on the “imagined” nation, Anderson (1991). On nationalism, see Gellner (1983). See Caramani (2004) on the process of political nationalization. 3 From Plato’s philosopher kings through Francis Bacon’s New Atlantis and Auguste Comte positivism up to present-day “guardianship”, such utopias have been recurrent. 4 Not coincidentally, these transformations are accompanied by the birth of social science and social and education policies. Max Weber (1922) specifically analysed the bureaucratic state (see Poggi 1978). 5 In today’s vocabulary, this tension is expressed in terms of responsive vs responsible government (Mair 2009) or democratic input vs effective output legitimacy (Scharpf 1999). 6 Caramani (2000) describes comparatively the processes of democratization and includes the historical literature. On different forms of political representation, see Manin (1997), Pitkin (1967) and Urbinati (2006). 7 It  is Manin (1997) who speaks of the “triumph” of the election over the lot and of a “democratic aristocracy”, itself an oxymoron capturing the duality of democracy/​efficiency. This dual democratic conception is the topic of Chapter 2 by Sánchez-Cuenca. 8 This  double role of political parties as bridging the tension between demands for responsiveness and demands for responsibility was first highlighted by Birch (1964) and Dahl (1956). It is the reason why Schattschneider spoke of modern democracy as “unthinkable save in terms of the parties” (1942: 1) and Sartori of citizens as “represented through and by parties” (1968: 471). Famously, this arrangement came to be known as the “responsible party model” (APSA 1950). 9 The  elitist critique is easiest to identify in theorist such as Michels, Ostrogorski, Mosca and Schumpeter. The populist critique goes back to Rousseau and Schmitt. Juxtapositions with pluralism can be found in Dahl (1956), O’Donnell (1994) and Riker (1982). More recently one finds it in Laclau (2005, see also Laclau and Mouffe 1985) and Rosanvallon (2011). 10 Major contributions on populism are Albertazzi and McDonnell (2008), Canovan (1999), Kriesi (2014), Mény and Surel (2002), Mudde (2007), Mudde and Kaltwasser (2012). Theoretically, on the populist conception of democracy, see Urbinati (1998, 2014b). 11 Literature on technocracy includes Centeno (1993, 1994), Dahl (1985), Dargent (2015), Fischer (1990, 2009), Meynaud (1969), Radaelli (1999b) and de la Torre (2013), among the many references cited throughout this Introduction. 12 These commonalities and differences are discussed in detail in Caramani (2017). 13 This is Easton’s phrase (1953). See also Lasswell’s logic of “who gets what” (1936). 14 Markets as instruments of collective decision making in the neoliberal and in the public choice critique of democracy as self-rule, are not considered as instances of technocracy (see Chapter 2 by Sánchez-Cuenca).

Introduction: the technocratic challenge  23 15 Also, in the perspective of public choice (the Virginia and Rochester Schools discussed in Chapter  2 by Sánchez-Cuenca), because the aggregation of preferences through democratic methods produces arbitrary results (Arrow 1951, Riker 1982). 16 Representation as acting in the interest of someone else is the definition of active representation by Pitkin (1967), as opposed to symbolic or descriptive involving a “being” rather than “doing”. The long-term goal of political action possibly involving prospectively also future generations refers to “responsibility” (Goetz 2014, Magaloni 2006, Thompson 2010). The holistic element is referred to in Pitkin as “unattached interest” (as opposed to subjective ones specific to each social group) or in Rehfeld (2009) as “good for all”. For an overview of the representation features of technocracy, see Caramani (2017: 60–61). 17 Terminologically, “technocracy” refers to power (and, by extension, to a form of political legitimacy and representation)—power of or through technical knowledge (financial, environmental, etc.)—as implied by the cracy in the term (although Tortola in Chapter 3 extends it to the power to decide on means, rather than goals). 18 This  aggregation results in representative plural democracies through electoral competition, interest representation and other forms of “input” towards the political system. The scepticism toward pluralism is a shared feature with populism, which is anti-elitist with a strong belief in the will of the people as opposed to scientific rationality. 19 See Akin (1977) and Elsner (1967) on the fortunes of technocratic movements in the United States. Soviets or theocracies or military regimes may be close examples (see Hoffman and Laird 1985 on technocratic socialism). Centeno (1994) applies this term to the Mexican state put in place during the Salinas “revolution” (Salinastroika) in the 1990s. 20 Defined as discourse (see below and Chapter 5), technocracy can become a tool used by actors (whether or not they are considered technocratic) in the same way in which non-populist actors can make populist statements. 21 Max Weber’s treatment of bureaucracy in the modern state is the classical reference. Mercantilism, population movements (urbanization) and industrialization increase the “cybernetic capacity” (Flora 1977: 114) of states to monitor societies through statistics as the “science of the state”. See Caramani (2000: 1005–15) on the origin of state statistics. 22 “War” (i.e. international competition) makes states, in Tilly’s famous phrase, as much as states make war. The role of competition in the development of the Western state is stressed by the world-systems theory from Wallerstein (1974: 348–49) through to Centeno (1994: 21–26), who applies it to Mexico. The argument extends to the need to “adjust” to, as well as compete within, supranational systems and therefore to rely on economic and legal expertise. 23 Scholars of the Industrial Revolution point precisely to territorial fragmentation to explain its origin in Europe rather than in imperial polities such as China. See, for example, Landes (1969). 24 Starting with Mair (2009), see also Bardi et al. (2014a, 2014b) for broad overviews of the consequences of parties losing touch with society by becoming close to the state, and of the increasing ideological similarity between them diminishing “choice” during elections. 25 Non-majoritarian institutions go under the label of type II governance (Hooghe and Marks 2003). On how presidents in Latin America try to navigate such complexities through technocrats, see Schneider (1998). 26 Not surprisingly, this period saw an increase in technocratic cabinets throughout Europe, especially as the financial crisis led to the intervention of teams (“troikas”) of international finance bureaucrats composed of the European Commission, the International Monetary Fund (IMF) and the European Central Bank, as in the

24  Daniele Caramani case of the Greek financial crisis. On the appointment of technocratic cabinets, see Marco Valbruzzi in Chapter 6 as well as Wratil and Pastorella (2018) and later in this Introduction. 27 To use Pitkin’s terminology (1967), such identification is “unattached”, i.e. not derived from the aggregation of individual preferences. Already, classic analyses of elites pointed to the depoliticization of the public sphere at large (Mills 1956). 28 In  sociology, too, one can observe such views from Saint-Simon, Comte and Karl Mannheim (on these thinkers, see Fischer 1990 and 2009) but even in ancient times with Plato and Aristotle. On non-majoritarian institutions, see Thatcher and Stone Sweet (2002). 29 See Bickerton and Invernizzi Accetti (2017), Bourdieu (2002), Laclau (2005), Laclau and Mouffe (1985), Pettit (2004) and Rosanvallon (2011) in regard to the depoliticizing tendencies of the welfare state and the market as “unifying logics” doing away with conflict. 30 This vision of parties does not include their capability to formulate visions and projects (albeit alternative ones) for the common good of the society. It is, rather, a corporatist vision of components working together, as in a machine or body. 31 Plurality in technocracy is reduced to the opposition between right (the experts) and wrong (the incompetents), such as in populism, in which it is reduced to the opposition between the virtuous people and the corrupt elite (see Caramani 2017: 64). 32 This is different for parties that have as their main function “articulation” rather than government, such as niche or single-issue parties. 33 This feature of the responsible party model is stronger in PR systems in which parties are more structured in terms of party discipline and cohesion (Birch 1964, Katz 2014) and are nationalized (Caramani 2004, 2015: 284–86, Pierce 1999, Pitkin 1967: 175). 34 One finds such elements in Robert Putnam’s (1977) study on civil servants’ mentality that includes objectivity, political neutrality and scientific view of policymaking. 35 Depending on the degree of ideological content of constitutions or treaties (or the degree to which one considers them as ideologically driven), the judiciary, too, loses its neutrality. One such case is the European Court of Justice (ECJ) whose rulings are based on the Treaties of the EU. Insofar as these include specific policies, the ECJ is not an ideologically neutral institution (Schmidt 2018). On neoliberal economic reforms in Latin America, see Arce (2006), Biglaiser (2002), Boylan (1998), Conaghan and Malloy (1994), Silva (1991) and Teichman (1997, 2002, 2004). 36 For a critique of the debate on “responsibility vs responsiveness”, see Goetz (2014). 37 On responsive technocracy, see Rauh (2016). On risk, see Rothstein (2006) and Rothstein et  al. (2006). On reputational risks in particular, and on accountability toward the public, the media and politicians, see Gilad (2015) and Maor and Sulitzeanu-kenan (2016). 38 “Unaccountable” is not  a great term, yet it is less misleading than “irresponsible” (although “responsibility” is sometimes used for “accountability”) (Sartori 1976: 19). Especially in parliamentary systems, the latter refers to a different dimension of representation (“good” and long-term governing) discussed above. 39 On anticipatory representation, see Mansbridge (2003). 40 The literature on political representation, since Burke’s discussion on the role of the presence of Irish Catholic representatives and American colonists in parliament, has addressed the influence of the “to be” on the “to do” (in particular, the function of supply of information). See Mansbridge (1999), Williams (1998) and Young (2000). 41 On the politics of presence vs politics of ideas see Phillips (1998), who refers, however, to the presence of disadvantaged groups and their diminished “social meaning”. 42 A  further dimension of representation, symbolic representation, is not  discussed here. For a short discussion, see Caramani (2017: 62–63) and also footnote 60 later.

Introduction: the technocratic challenge  25 43 On technocrats as agents or actors, see Dargent (2015: 14–19). 44 This is a more general view on state and bureaucratic autonomy (Skocpol 1985: 9). Of course, it is problematic to speak of “preferences” in the case of experts whose positions are claimed to be shaped by rational analysis. 45 Instrumentalist explanations of the role of technocrats think of experts as “tools”. Strictly speaking, however, following the definition above, agents, technocratic bodies and personnel cease being “technocratic” as they do not allocate values but merely execute as agents. They remain at the level of technicians and experts without power. By definition, technocrats are actors, who formulate and pursue goals, independently of groups, and the source of judgement is personal, based on expertise. 46 Such levels are linked. Identifying who qualifies as a technocrat is necessary to be able to establish how significant the technocratic component is of a given cabinet (Chapter  6) and its impact on policy and behaviour in negotiations (Chapters  7, 8 and 10), In turn, characterizing an institution or an actor as technocratic helps the interpretation of its discourse (Chapter 9), and allows assessment of the electorate’s support for it (Chapters 11 and 13). 47 On the rational function of states, see Birnbaum (1988) and Badie and Birnbaum (1983), not coincidentally having the strong centralized French state in mind. Here rests also the paradox that, while often championing neoliberalism, technocrats need a state to implement reforms (on the “orthodox paradox”, see Kahler 1990, 1992). 48 See Eduardo Dargent’s Chapter 12 for a review of regimes in which technocrats are active in Latin America. The argument is that técnicos (as “experts” are termed in the Latin American literature) become essential to políticos because of their expertise in complex matters, but also that experts use their knowledge to actively build their own autonomy. The cost of getting rid of them thus increases continuously. 49 On the more specific negative effects of autonomous and, ultimately, incompatible technocrats on the quality of democracy in the 1990s see Centeno (1993), Centeno and Silva (1998), Conaghan and Malloy (1994) and Teichman (1997). Dargent’s analysis (2015) confirms the protracted power of non-elected actors over democratically elected politicians and the serious shortcomings of Latin American democracies. 50 It is controversial whether or not the judiciary can be considered a technocratic body. It is, insofar as courts influence the policy process and the allocation of values. This is the case in highly “judicialized” contexts in which constitutional review and rights adjudication becomes an extension of the policy process and an amendment to legislation (see Sadurski 2013, Shapiro 1964 and Shapiro and Stone 1994). Proof of its technocratic nature is also the attacks from populists speaking of courts as “enemies of the people”. 51 On these cases, see Brunclík (2015), Marangoni and Verzichelli (2015), Pastorella (2014, 2016) and Pasquino and Valbruzzi (2012). 52 Different languages label such individual actors differently. The  influential Latin American literature speaks of técnicos (or tecnócratas, as opposed to políticos); similarly, Italian would use tecnici (or tecnocrati). 53 This is the difference with “technopols”, as presented in Chapter 7 by Alexiadou. See also Alexiadou and Gunaydin (2019) and Joingnant (2011). 54 Collier’s definition of a technocrat captures these two elements: “[i]ndividuals with a high level of specialized academic training which serves as a principal criterion on the basis of which they are selected to occupy key decision-making or advisory roles in large, complex organizations—both public and private” (Collier 1979: 403). 55 One finds this function in the responsible party model cited earlier, but also in later literature speaking of “partisans’ skills” (Rose 1969) and “officials’ recruitment” (Sartori 2005, Katz 1987, Mair 2008).

26  Daniele Caramani 56 This  book covers one such supranational organization in particular—namely, the EU: see Chapter 11 by Costa Lobo and McManus on the legitimacy of EU policies, and Chapter  4 by Van der Veer on the EU’s responsiveness address this level of governance. 57 This section focuses on discourse, but includes also actions that may be defined as technocratic (for example, a non-responsive decision against voters’ preferences in the name of responsible policy). 58 The relationship between discourse and ideology is not addressed here. 59 For a comparison on the two types of quantitative content analysis, see Rooduijin and Pauwels (2011). 60 Brazil’s flag mentioning “order and progress” reveals the republican project of 1889, imbued with positivist references to government by an enlightened elite acting on the basis of scientific methods to bring about industrialization and social welfare. The  hierarchical, corporatist view of society influenced other countries in Latin America as well—in particular Mexico—with thinkers such as Francisco G. Cosmes. 61 The  discourse extends to single representatives, interpreting their representative “roles” in different ways, more as trustees than as delegates of the voters. One would expect populist representatives to see their role as translating the people’s will, and technocratic ones to be less responsive to that will. 62 On populist attitudes, see Akkerman et al. (2014, 2017), Castanho Silva et al. (2019), Oliver and Rahn (2016), Schulz et al. (2017) and Van Hauwaert and Van Kessel (2018). 63 The survey items are: (1) “It would be better for the country if politicians stopped talking and concentrated on solving actual problems”; (2) “compromise in politics is really selling out one’s principles”; (3) “this country would run better if political decisions were left up to successful business leaders”; (4) “this country would run better if political decisions were left up to experts instead of politicians and citizens” (Hibbing and Theiss-Morse 2002). For  a recent example along the same lines, see Ruiz-Rufino and Alonso (2017). 64 The  strength of the challenge and the bleaker view of the state of representative democracy is accrued when academics claim that democracy is the rule of the ignorant and the irrational, even if this echoes classical concerns by Plato and Aristotle (Brennan 2016). Such views are critical of citizens’ competence and point out that political participation and democratic deliberation tend to make people more irrational and biased. Epistocracy is invoked as the “rule of the knowledgeable” to replace citizens’ democracy. 65 This is the position of the radical approach to democracy. Laclau and Mouffe (1985) in particular argue in favour of antagonism and contestation as a corrective to cartelization, ideological convergence, constitutionalism, consensus, the market and welfare administration. For an overview, see Rovira Kaltwasser (2012). 66 The case of the EU is particularly relevant, and similar to other cases of consociational democracies, given the binding nature of its directives for the member states. This makes it different from other supranational organizations such as the UN or the IMF. As with all multicultural polities, it is one in which majoritarianism cannot be applied given the ascriptive nature of cleavages between nations. 67 On these points see Dickson (1981), Habermas (1970) and Offe (1973).

Part I

Concepts and theory

1 Technocracy and political theory Christopher Bickerton and Carlo Invernizzi Accetti

Introduction: a critique of classical and contemporary arguments The notion of technocracy is at the heart of contemporary political discourse and debates. At the height of the Eurozone’s economic crisis, in November 2011, both the Italian and the Greek prime ministers were replaced by figures widely identified as technocrats (Donadio 2011, Verney and Bosco 2013). In  debates over the European Union’s supposed “democratic deficit”, its institutions are often described as technocratic (Føllesdal and Hix 2006, Habermas 2015, Majone 1998, Moravcsik 2002). Finally, in the discussions that preceded the 2016 vote for “Brexit”—but also in many of the national elections that have taken place since then, such as in the United States and in France—the choice has often been framed in terms of an opposition between nationalist “populism” on the one hand and liberal “technocracy” on the other (Bickerton 2016, Bickerton and Invernizzi Accetti 2015, Caramani 2017, Giugni and Rubinelli 2015, Invernizzi Accetti 2016, Leonard 2017). Underpinning most of these references to technocracy is the assumption that the latter consists of an appeal to expertise as the basis for political legitimacy. Academic definitions of the term reflect this assumption. In his seminal 1969 text on the subject, Jean Meynaud defined technocracy as “a system of government in which technically trained experts rule by virtue of their specialized knowledge and position in dominant political and economic institutions” (Meynaud 1969: 31). Along similar lines, in a more contemporary example, Duncan McDonnell and Macro Valbruzzi laid out three requirements for classifying a prime minister or a minister as a technocrat (see also Valbruzzi, Chapter 6 in this volume, on technocratic cabinets). The first was that, at the time of appointment, he or she had never held public office under the banner of a political party. The second was not being a formal member of any party and the third was to have “recognized non-party political expertise, which is directly relevant to the role occupied in government” (McDonnell and Valbruzzi 2014: 657−58). From such definitions, it follows that there is an irreducible tension between technocracy and democracy. Democracy rests upon a principle of “identity” between the rulers and the ruled, where the “people”—construed as the object of

30  Christopher Bickerton et al. government—are also its subject and are therefore the ultimate basis of political legitimacy (Bobbio 2005, Dahl 1956, Habermas 1989, Kelsen 1929). In contrast, technocracy introduces a quite different principle of legitimacy. The term implies that some people can legitimately rule over others, by virtue of their technical competence or administrative expertise (Centeno 1993, Centeno and Silva 1998, Dargent 2014, Fischer 1990, 2009, Radaelli 1999b). In this chapter, we do not deny this tension between technocracy and democracy. However, we find that defenders of technocracy today rarely make this tension explicit. Instead, a claim about the complementarity between technocracy and democracy underpins contemporary arguments for technocracy. Technocratic political arrangements are said to bolster a broadly democratic system of government by increasing its “output legitimacy” in specific policy domains (Scharpf 1999). Thus, Claudio Radaelli has drawn a useful distinction between what we call in this chapter the classical arguments for technocracy, those which “advocated for a direct rise to power of experts”, and more contemporary ones, which are “formally respectful of democratic values and institutions” (Radaelli 1999b: 24). The  intention behind these contemporary arguments for technocracy is to carve out spaces for (relatively) autonomous decision making by experts, within a political system that continues to draw its legitimacy from the democratic ideal of “identity” between the rulers and the ruled. The  widespread proliferation of “quasi-autonomous agencies” devoted to the regulation of specific areas of policy within the nation states is an example of this (Elgie 2006, Gilardi 2002, Rosanvallon 2011). So too is the delegation of specific policy competences to supranational bodies such as the European Central Bank or the European Commission (Bickerton 2012, 2016, Majone 1994, 1996, Moravcsik 2002). In  this chapter, we argue that despite the often-made claim about complementarity, contemporary arguments for technocracy continue to represent an important challenge to the democratic idea of collective self-rule. They do this by implicitly assuming that the boundaries between the democratic and technocratic areas of decision making are themselves matters of technical expertise. They thereby ultimately end up placing experts—and not the “people”—in the position of sovereign. As a result, the relationship between democracy and technocracy poses once again the question of where ultimate authority resides within a polity. Contemporary claims around complementarity between democracy and expertise elide this more fundamental conflict, which has been an integral part of the development of modern societies and their reliance upon expert knowledge. As Harald Laski put it back in 1931, in a pamphlet entitled “The Limitations of the Expert”, experts should be “on tap, not on top” (Laski 1931). This principle is far easier to formulate in theory than it is to implement in practice. The question of where final authority resides is constantly posed anew by concrete developments and policy innovations, as is clear from the discussion of the EU’s democratic deficit later in the chapter. To demonstrate this, the chapter proceeds in three parts. In  the first, we reconstruct the “classical” argument for technocracy, through an engagement

Technocracy and political theory  31 with the work of the author we think has formulated it in the most powerful and long-lasting way: Plato. We show that his argument for empowering philosophers relied on a conflation of two domains that classical Greek political culture had previously kept rigorously separate: that of politics, construed as a domain of “self-realization”, and that of economics, construed as the domain of the satisfaction of material “needs” (Arendt 1958). This conflation allowed Plato to portray politics as a techne—that is, an “art” or a “craft” requiring a specific kind of knowledge. On this basis, Plato could maintain that only those with such knowledge (i.e. those he called philosophers, and what we would probably refer to as experts) should be entitled to rule, independently of whether they obtain the consent of the demos. In  the second part of the chapter, we show that while versions of this “classical” argument for technocracy persist in some circles today, the transfer of power to experts is predominantly justified through an idea of a separation between different policy domains. Whilst some domains should be left to the “people” to decide over democratically, others are, for a variety of reasons, better entrusted to expert decision making. Far from representing an alternative to democracy, technocracy thus complements and enhances democratic legitimacy. In this second part, we show that this argument comes in a variety of guises, and we identify several different criteria that have been put forward for determining how the boundaries between technocratic and democratic policy areas should be drawn. However, we also observe that these different accounts share the common assumption that the boundaries between separate policy areas are themselves a matter of technical expertise, and should therefore not to subject to the dynamics of political contestation. In this way, contemporary arguments for technocracy end up subordinating the democratic exercise of collective self-government to a prior exercise of technical competence, which subverts the chain of authorization commonly associated with democratic politics. We can also see here a clear difference between contemporary instances of empowering technocratic authority and what is more generally referred to as “liberal constitutionalism” (Elster and Slagstad 1988, Holmes 1988). Whilst the latter is concerned with empowering non-majoritarian institutions, these institutions were created as part of a wider constitutional settlement bound up with attempts at enacting popular sovereignty. As Martin Loughlin has put it, the non-majoritarian components of liberalism are best thought of as expressions or explications of the basic principle of popular self-government rather than as limitations or external constraints on this principle (Loughlin 2003). Contemporary instances of delegating power to technocratic institutions are not  part of any such constitutional moment, and their legitimacy rests far more heavily on assumptions about the specific characteristics of certain policy domains.1 The  same applies for distinctions between technocracy and judicial power. There  is an element of expertise associated with law, and the practice of courts requires that conflicts are depoliticized and both parties accept the arbitration of the third party (Bickerton 2017, Carr 1995). However, the law itself originates in the deliberation and decisions of elected bodies, such that

32  Christopher Bickerton et al. the boundary between judicial power and technocracy more broadly is clear. As Anthony King put it, judges have become controversial in recent years not because they are seeking to expand their power at the expense of a parliament or an executive, but because politicians have invited judges to make controversial decisions (2015: 273). The third part of the chapter makes the case that this contemporary way of arguing for technocracy is more insidious for democratic political systems than the classical arguments advanced by Plato and his successors. The Platonic argument for technocracy construes it as a utopian ideal, which can function as the ground for a partisan political project within a democratic political regime (as was the case, for instance, with the “Technocracy” movement of the 1930s in the United States). In contrast, contemporary arguments for technocracy undermine the very space in which a public contestation of the relative merits of technocracy and democracy could take place, by defining a priori the objective boundaries of their respective spheres of legitimacy.

The classical argument for technocracy It is generally accepted that Plato’s argument for the idea that philosophers ought to rule relied on a prior conception of politics as a techne: a term now commonly translated as “art” or “craft”, but which has a broader significance in the context of Plato’s work (Harvey 2009, Wadia 1987, Wild 1963). Its two key features are the idea of an activity oriented towards a specific goal (ergon) and the assumption that this requires a specific kind of knowledge (episteme). The most common examples Plato refers to are medicine—the goal of which is supposed to be “health”, and the knowledge required to attain it, “knowledge of the human body”—and what he calls “shepherdship”, namely the herding of sheep, which requires knowledge about the “nature of sheep and their needs” (Bambrough 1963). Analogously, Plato assumed that the goal of politics is the “herding of human beings” (The Statesman: 267c). The knowledge required for that was the “knowledge of human nature” and its “needs”. On this basis, Plato reaches the notorious conclusion that philosophers (i.e. people who possess a specific kind of knowledge and therefore skills) ought to rule. In the dialogue on The Statesman, for instance, the main character asserts that: “Physicians offer a particularly good example of this point of view. Whether they cure us against our will or with our will, by cutting us or burning us or causing us pain in any other way … we call them physicians just the same, so long as they exercise authority by art or science” (293b). By analogy, the same character then adds, “among forms of government, that one is preeminently right and is the only real government, in which the rulers are found to be truly possessed of science, not merely to seem to possess it … whether their subjects are willing or unwilling” (293c). This is the core of what we shall call the classic argument for technocracy. What still needs to be explained, however, are the grounds on which Plato maintains that politics can be understood as a techne. Some commentators have suggested

Technocracy and political theory  33 he simply “takes this for granted” (e.g. Accattino 1997: xv), but that would mean his whole argument for technocracy was just question-begging. In reality, Plato presents a powerful argument to support his key premise. In The Republic, when Socrates and Adeimantus begin constructing Plato’s “ideal city”, Socrates makes clear that the reason why human beings enter into cities in the first place is the satisfaction of their material “needs”. “The origin of the city”, he states, lies in the fact that we are not, any of us, self-sufficient … Different individuals, then, form associations with one person to meet one need, and with another to meet a different need. With this variety of wants, they may collect a number of partners and allies into one place of habitation, and to this joint habitation we give the name city (269b–c). This premise represented a radical break with key features of ancient Athenian political culture, which Plato was writing in (and against). Both Hannah Arendt (1958) and Moses Finley (1973) have emphasized that classical Greek political culture distinguished categorically between the domains of the oikos—the “home” construed as the domain for the satisfaction of “material needs” (i.e. production and reproduction)—and the polis, construed as the domain where free men would meet and act together as equals in pursuit of their “self-realization”. From this perspective, “economics” was understood as the set of norms (nomos) that pertained to the oikos. As Arendt further notes, these are necessarily “despotic” norms since the satisfaction of material needs is a techne which admits of right and wrong approaches. In contrast, the polis was assumed to be a domain in which “free men” would encounter each other on a basis of equality, and collectively strive for their individual “self-realization” (Arendt 1958: 28−37). By claiming that human beings originally entered into cities to satisfy their material needs, Plato radically broke with this whole framework, implying that politics was about what his contemporaries would have called economics. This is precisely what Aristotle accuses him of at the beginning of his own treatise on Politics (1252a: 1−7). The conflation between the domains of politics and economics allows Plato to maintain that politics is a techne and therefore that only people who possess a specific kind of knowledge should rule. As we pointed out earlier, the Greeks assumed that the satisfaction of material needs was a science, and hence required a specific kind of knowledge to be carried out correctly. In contrast, the distinctively political activity of encountering others as equals and acting together for the purpose of everyone’s self-realization does not suppose any specific knowledge that would be available to some more than others. This is expressed, in the form of a myth, in Plato’s Protagoras, where the title character attempts to defend democracy against Socrates’s competence-based approach. He maintains that whereas other “skills” were originally distributed by the gods in different measures to different animals and human beings, the capacity of “being political” was distributed equally to all human beings (320d1−322a2). The  deeper point is that if politics is construed as a domain of freedom, and as oriented towards the “self-realization” of its participants (as opposed to the

34  Christopher Bickerton et al. satisfaction of their material “needs”), then there cannot strictly be a science of politics. Human beings can only be free if they are allowed to pursue their selfrealization in the way they think best. It is in this sense, therefore, that we contend that Plato’s argument for construing politics as a techne depends on a prior conflation of the domains of politics and economics. Doing so excises the notion of freedom from the political domain, and thus reduces the latter to the exercise of a science oriented to the satisfaction of material needs. This is made explicit in Plato’s dialogue on The Statesman, where at one point the main character asks: “Are ‘statesman’, ‘king’, ‘master’ and ‘householder’ one and the same, or is there a science answering to each of these names?” (358e). He soon answers himself: “A large household may be compared to a small state”, from which it follows that: “There is one science of all of them, and … this science may be called either ‘royal’ or ‘political’ or ‘economical’—we will not quarrel about the name” (359c). Plato goes on to establish a distinction between the “true statesman” and the “mere politician”, which Sheldon Wolin has captured by focusing on the different “constituency” each is supposed to represent: The popular ruler owes his power to an ability to sniff the moods and aspirations of the populace, to juggle a wide variety of variables and strive for an ad hoc solution … In contrast, the Platonic ruler has a different constituency, for he is not first and foremost a “political man”, but a philosopher endowed with political power … In following the true art of statesmanship, [he] conforms to the knowledge made possible by philosophy and discharges the obligation of the philosopher to pursue truth. And in following the dictates of his art, rather than the wishes of the community, the philosopher qua ruler satisfies the demands of rulership, because the ends of his art coincide exactly with the interests of the community (Wolin 2004: 42−43). If we substitute the term “expert” for that of “philosopher” in the previous passage, we obtain a very clear statement of what is still widely held to distinguish the technocrat from the professional politician in contemporary language. Whereas the accountability of the latter is primarily to the “wishes of the community”, the former is first and foremost accountable to “truth”. This distinction was originally intended by Plato to function as an argument against democracy, on the grounds that politicians accountable to the “wishes of the community” could not also fulfill the primary function of “true statesmanship”. At the heart of the classical argument for technocracy, therefore, lies an opposition between (subjective) freedom and (objective) truth.

Contemporary arguments for technocracy Whereas the Platonic argument for technocracy remained the backbone of political common sense for many centuries after Plato, the strategic context for technocratic critiques of democracy has changed in the modern democratic age. Some explicitly anti-democratic arguments for technocracy have been advanced in this

Technocracy and political theory  35 period, for instance by the so-called “Technocracy” movement that developed in the United States in the early years of the Great Depression (Akin 1977, Elsner 1967). Echoes of this remain in some strands of “technological solutionism” very popular in and around Silicon Valley (Kling 1996, Morozov 2014, Schmidt and Cohen 2013, Segal 1985, Thiel 2014, Turner 2006). By and large, however, the idea that technocracy should replace democracy has become rather marginal. The political project of transferring power away from democratically elected “politicians” and towards presumptively more competent “experts”, however, is far from having seen its day. Instead, contemporary arguments for technocracy have taken a new form. There are a variety of examples to draw on, from philosophical inquiries into “epistocracy” to the extensive writings on the “politics of expertise” in science and technology studies (Collins and Evans 2002, Estlund 2008, Fischer 2000, Holst 2012, Jasanoff 2012, de Vries 2016). In this chapter, we explore contemporary arguments for technocracy by looking at ongoing academic discussions of the European Union’s purported “democratic deficit”. In parallel to the dispute over the democratic qualities of the European Union itself, there has also developed a significant body of literature arguing that the latter’s purpose is to complement the exercise of democratic self-government at the national level. This is achieved by “insulating” specific areas of policy from democratic contestation, and thereby allowing trained experts to regulate them in a more competent way (e.g. Majone 1998, Monti and Goulard 2012, Moravcsik 2002). This body of literature accepts the charge laid against the EU that it is technocratic, but argues that this enhances the quality of democracy in Europe as a whole. Whilst the EU as a political system is different from what exists at the national level, there are good reasons to focus on the EU’s “democratic deficit” debate in order to examine the relationship between technocracy and democracy. One reason is that in the light of the Eurozone crisis—and the experience with technocratic governments in Italy and Greece—there has been a great deal of scholarly and political discussion around this issue, where the focus has been precisely on identifying the legitimacy (or not) of technocratic administrations (e.g. Bertsou and Pastorella 2017, Pastorella 2016). In  light of these debates, it is very useful to return to some of the seminal writings on the EU’s “democratic deficit”. A second reason is that whilst the EU’s political system is distinct, it is at the same time firmly embedded within national political structures, and particularly the constitutional arrangements of the EU’s member states. The arguments that are used in order to support the delegation of authority to technocratic bodies at the EU level are symptomatic of a more general understanding of how democracies relate to the phenomenon of unelected power (Tucker 2018). It is therefore possible to apply some of the conclusions we make regarding the problems of the EU’s “democratic deficit” to the more general debate about the relationship between technocracy and democracy. Indeed, particularly in respect of the delegation of policymaking authority to expert bodies, the expansion in the number of delegated agencies at the EU level is outmatched by a similar phenomenon the national level, the so-called “agentification” of national states (Pollitt 2015).

36  Christopher Bickerton et al. In  an article entitled “In  Defense of the Democratic Deficit”, Andrew Moravcsik has written that: “Within the multi-level governance system prevailing in Europe, EU officials enjoy the greatest autonomy precisely in those areas … in which many advanced industrial democracies, including most Member States of the EU, insulate themselves from direct political contestation” (Moravcsik 2002: 613). On this basis, he reaches the seemingly paradoxical conclusion that: “To the extent that this is so, the EU may be more ‘democratic’ precisely because it is, in a narrow sense, less ‘representative’” (2002: 614). Here, the tension between technocracy and democracy at the heart of Plato’s work is eliminated entirely. Instead, we find that as long as the power of experts is confined to specific policy areas, then it will enhance democratic self-rule. The transfer of power to experts is no longer justified on the basis of a conflation between the oikos and the polis but rather with reference to a distinction between policy domains. Some are dealt with by “democratic” means, others by “technocratic” means, with an overall gain for the quality of democracy. This  distinction between policy domains is drawn in a variety of ways. What all accounts have in common is the assumption that the boundary-drawing exercise is itself a technical matter, requiring expert evaluation. In the remainder of this section of the chapter, we will detail three attempts at creating a framework for how one should distinguish between “democratic” and “technocratic” policy areas. The third section will discuss critically these different arguments. In  his work on the “democratic deficit” of the EU, Andrew Moravcsik provides three normative arguments for “insulating” certain policy areas from “direct contestation” (2002: 613). The first is based on the assumption that ordinary citizens are “rationally ignorant” of issues where the cost of acquiring the necessary information to be able to participate competently in policy discussions exceeds the benefits that can be gained from such participation. From this, Moravcsik infers that it is “rational” for citizens to delegate decision making to trained “experts” in domains such as environmental policy and food and drug administration (2002: 614). Secondly, Moravcsik argues that the delegation of political power to “independent bodies” is necessary to protect minorities from the “tyranny of the majority”, highlighting in particular domains such as “administration of justice” and the safeguarding of “individual rights” (2002: 614). Thirdly, Moravcsik argues that the “insulation” of collective decision making may be necessary to protect against the risk of capture of the democratic process by “special interests”. He  provides the example of trade policy, where well-organized protectionist interests may “win out” against the more diffuse pro-free trade interests of consumers and the economy as a whole (2002: 615). As they are not accountable to immediate electoral constraints, “independent authorities” can take into account the broader interests of the “collectivity as a whole” and thus guard against this risk (2002: 615). Another highly influential participant in the ongoing debate on the EU’s “democratic deficit” is Giandomenico Majone. In a number of separate contributions,

Technocracy and political theory  37 Majone has argued that the “counter-majoritarian” character of the European integration process should be understood by situating the EU within a broader theory of the “regulatory state” (Majone 1994, 1996, 1998). This  notion relies on a distinction between “Pareto improving” and “redistributive” policies. The former are assumed to be policies that make “some people better off, without making anyone else worse off”. The  latter “allocate resources between social groups” and therefore inevitably make some people better off at the expense of others (1994: 65). The  hallmark of the regulatory state, of which the EU is a prime example, is its preoccupation only with “Pareto improving” policy areas. Majone argues that “redistributive” policies inevitably generate “conflicts of interest” between social groups, and therefore require “political” and “democratic” (he uses the terms interchangeably) procedures to be legitimately decided upon and implemented (1996: 124). By contrast, “Pareto improving” policies can in principle form the object of a “consensus” amongst all the affected parties. For this reason, entrusting them to “independent regulatory bodies” poses no normative problems in Majone’s view (1996: 125). In a similar line of reasoning to Moravcsik, Majone suggests that the specific areas of competence entrusted to the European Union broadly correspond to those that ought to be entrusted to “independent regulatory bodies” because of their Pareto improving character. The European Union’s purported “democratic deficit” is therefore not a problem, as “regulatory bodies” need not  conform to the same democratic standards as redistributive policies (1998: 18−19). Before considering the problems with these arguments on technocracy in more detail, it is worth noting that neither Moravcsik nor  Majone take into account the possibility that significant political disagreements may exist over the criteria they use to determine which policy areas should be dealt with democratically or through expert deliberation. According to Majone, regulation should be “used to increase the efficiency of the economy by correcting for market failures” (1994: 82). However, he admits that it is possible for governments to opt for public ownership instead. In those instances, the goals are no longer just market correcting, they also involve substituting market criteria for “political and administrative ones” (1994: 82). The goal of regulation is therefore to correct market failures by creating a regime of monitoring, control and rule-setting staffed by those with the required expertise. However, the prior decision to pursue a regulatory route lies in the background, as a permissive or operative condition for the development of the regulatory state in the first place. Majone observes that from the 1930s through to the 1980s, the preferred path in parts of Western Europe was public ownership and not regulation. The decision on what counts as a regulatory or a redistributive policy area is therefore not simply a matter of measuring up a policy area to a set of objective criteria. It is also the result of deciding, for reasons of politics and ideology, to pursue one route rather than another. Crucially, this poses the question of who shall make a decision such as this. When Majone describes the shift towards regulation as a “redrawing of the

38  Christopher Bickerton et al. borders of the public sphere in a way that excludes certain fields better left to private activity” (1994: 80), he says nothing about who is doing the redrawing. He writes using only the passive voice, whereas it is obviously him (that is, an expert) that is drawing the boundary. When discussing the “distinctive institutional competence” of counter-majoritarian institutions, he argues they should exercise “self-discipline” when deciding on the policy areas they would like to be involved in (2009: 169). The policing of boundaries between majoritarian and counter-majoritarian institutions is thus left to the “self-restraint” of the independent experts running these institutions. Arguments such as these suggest that while Majone argues for a “competence-based division of labor” between technocracy and democracy, the boundary between the two is controlled in the final instance by experts. A final argument for the complementarity of democracy and technocracy considered here is that put forward by Mario Monti and Sylvie Goulard in their 2012 treatise on Democracy in Europe. Reflecting their preoccupation with the Eurocrisis, which was at its peak as they were writing (and was of central political relevance in Mario Monti’s nomination as Italian prime minister in November 2011), Monti and Goulard stress the importance of temporality in politics. Democracy, they insist, is the bedrock of all political legitimacy. However, it is also subject to the “tyranny of the short term”. Since elected politicians inevitably operate within the time frame of periodic elections, their temporal horizon can never extend far into the future (2012: 37−38). Unelected experts, on the other hand, are able to play a longer game and make decisions that take into account future possibilities and uncertainties. Monti and Goulard call them “priests of the long term” (2012: 40). On this basis, Monti and Goulard take a stand against what they call the “demonization of expertise”. In their view, what is needed is a “correct combination of technocracy and democracy” in order for political action to be “reactive in emergencies” and able to “anticipate the long term” (2012: 41). The relationship between technocracy and democracy is posed here in terms of the search for an appropriate “balance” between them: one cannot be entirely sacrificed to the other as both are necessary to ensure that the political system achieves what Monti and Goulard call “temporal pertinence” (2012: 41). Once again, however, one may ask how exactly the conflicts of competence can be resolved when framing a problem as short- or long-term has important distributional consequences. For instance, an example Monti and Goulard repeatedly refer to in their analysis is the delegation of monetary policy in the Eurozone to an “independent” central bank. This is justified on the basis of the claim that all economic agents have a “long-term interest in price stability”, whereas elected governments may have an incentive to manipulate interest rates in order to obtain “short term electoral advantages” (Monti and Goulard 2012: 39).2 It is not selfevident, however, that all economic agents have a “long term economic interest in price stability”. In  a society in which there are both debtors and creditors, the former clearly stand to benefit from a decrease in the real value of nominal amounts, whereas the opposite is true of the latter. And even if the claim is that, in the very long term, both have interest in the sustainability of the credit

Technocracy and political theory  39 system itself, that has proved to be compatible with a range—both positive and negative—of rates of inflation, whose distributional consequences are very different (see also Sánchez-Cuenca, Chapter 2 in this volume).3 Like Moravcsik and Majone, Monti and Goulard simply take for granted that the boundaries between different policy areas can be defined in a purely “objective” and “value-neutral” way. In fact, as we have suggested, it is sufficient to raise just a few questions about the specific way in which they propose to allocate competences to technocratic and democratic institutions to realize that this is a deeply political and normatively loaded decision. Other scholars have taken up certain aspects of this problem, challenging in particular the distinction between informative and distributive aspects of political decision making that is central to epistemic and deliberative democratic theory (e.g. Landwehr 2009, 2010), but without analysing specifically the relationship between technocracy and democracy. Even though the authors discussed above claim only that they are complementing democracy with a technocratic appeal to expertise in certain policy domains, they are in fact subordinating the democratic exercise of collective selfgovernment to a prior exercise of technical competence. This is done by denying the properly political character of the decision over how to draw the boundaries between different policy areas. In the accounts discussed herein, the responsibility for making this fundamental political decision about boundary delimitation is left to experts. Though the form of the defense of technocracy is different, in substance arguments about the complementarity between democracy and technocracy turn out to be surprisingly similar to the Platonic argument detailed in the first section of the chapter.

Towards a critique of contemporary arguments for technocracy The fact that contemporary arguments for technocracy make a formal show of deference towards democracy (by presenting themselves as arguments for reconciling the two) can be seen as an implicit endorsement of the normative value of the democratic ideal of collective self-government, if only at a symbolic level. Upon reflection, however, this indirect way of arguing for technocracy appears more of a challenge for democracy than the classic argument for technocracy we outlined in the first part of this chapter. We first see this when we inquire into the specific conception of democracy that contemporary advocates of technocracy want to reconcile expertise with. The notion of “Madisonian” (or “neo-Madisonian”) democracy has been at the heart of the work of Moravcsik, Majone and many others (Héritier 1999, Keohane et al. 2009, Nicolaïdis 2001, Zielonka 2006). In making the case that the European Union’s alleged “democratic deficit” is not as much of a problem as many believe it to be, Majone writes that “[w]hile the pure majoritarian model of democracy is opposed to any delegation of powers to non-majoritarian institutions, the pluralist or Madisonian model—which aims to share, disperse, limit and delegate power—provides a much more favorable constitutional environment for the development of such institutions” (1998: 18). The  notion of “Madisonian”

40  Christopher Bickerton et al. democracy is used here in opposition to the idea of a “purely majoritarian” conception of democracy. Its key feature is the “division” and “delegation” of powers to a “plurality” of competing political institutions, whose functions are to reciprocally “limit” and “complement” each other (1998: 18). This conception of democracy is more “favourable” to the delegation of power to non-majoritarian institutions principally because technocratic institutions such as independent regulators can be construed as a subset of the multiple competing political institutions created and set in competition with one another by the “Madisonian” system of government. This inscribes them within a broadly democratic horizon even if, strictly speaking, their function is only to “limit” and “complement” the democratic exercise of collective self-government. Indeed, some authors have theorized expertise as a new “pillar” or “branch” of democratic government, along with the judicial, legislative and executive branches (Vibert 2007). There remains, however, a crucial difference between a genuinely Madisonian conception of democracy and the argument made by authors such as Majone, Moravcsik, Monti and Goulard, who aim to reconcile democracy and technocracy in the name of building a new separation of powers.4 For Madison, the ultimate bedrock of political legitimacy remained the democratic ideal of “identity” between the rulers and the ruled (Bickerton 2011, Wood 1998, 2003). This is clear from the first line of the American constitution, which Madison contributed to drafting and which famously identifies the constituent power as “We, the People”. It is also clear from a passage in his Federalist essay n. 39, which defines a republic as “a government which derives all its powers, directly or indirectly, from the great body of the people” (Ball 1987: 255). As Madison explained in Federalist n. 46, “[t]he Federal and State governments are in fact but different agents and trustees of the people, constituted with different powers, and designed for different purposes …  [Whereas] the ultimate authority, wherever the derivative may be found, resides in the people alone” (Ball 1987: 297). In contrast, contemporary arguments for technocracy subordinate the democratic exercise of collective self-government to a prior exercise of technical competence, by treating the drawing of the boundaries between the policy areas as an objective and value-neutral exercise. Their authors end up placing themselves— as presumptively neutral “experts”—in the position of sovereign, instead of the “people” as a whole. This  challenges what Madison took to be the core principle of “Republicanism” and makes it implausible to imagine that contemporary arguments for technocracy are compatible with democracy, even in a broadly “Madisonian” sense. The  challenge to democracy stemming from contemporary arguments for technocracy is also visible if we refer to the Rousseauian distinction between the “sovereign” and the “government”, which Madison alludes to in the passage from Federalist n. 46 we quoted earlier. The sovereign, for Rousseau, is the ultimate bedrock of all political legitimacy, and for him this prerogative can only lie in the “people” as a whole (Rousseau 1997: 51−53). The government, by contrast, is defined as an “intermediate” body between the people and the state. Its

Technocracy and political theory  41 purpose is to “enact” the sovereign’s decision (1997: 82−86). As such, Rousseau maintains that it is quite right for the government to be composed of people with the “ability” to enact the sovereign’s decision—that is, “experts” in today’s language. Indeed, this is the reason why he famously maintains that democratic government is “not suited to men”, since that would require a “people of Gods” (1997: 92). His point, however, is that the sovereign (i.e. the people) must remain superordinate to the government, since it is entrusted with the responsibility of making the most fundamental political decisions. The  latter—namely, the government—only has the task of “enacting” them. If we compare this with the picture that emerges from contemporary arguments for “reconciling” technocracy and democracy, we can see how the latter effectively end up placing experts in the position of sovereign and the “people” as a delegated (i.e. governmental) power. Indeed, the power to determine the allocation of competences to different government bodies is frequently used as a short-hand definition of sovereignty, which in the arguments analysed here lies in the power of experts, not  the “people”. Contemporary arguments for “reconciling” democracy with technocracy thus end up delivering to the experts the mantle of sovereignty, and the “people” play the role of a subordinate power. It is left to the experts to decide what policy domains are appropriate for the application of the rules of electoral competition. From a democratic perspective, this appears to be a more insidious way of arguing for technocracy, compared to the “classical” argument for technocracy. Whereas, Plato presented his arguments in the form of a dialogue, opening up his claims to some semblance of public contestation, contemporary arguments for technocracy undermine the scope for legitimate contestation over this fundamental political issue by presenting their distinction between different areas of policy as if it were objective and value-neutral. Moreover, Plato’s defence of technocracy contained a “utopian” vision of overcoming the problems and limitations of political conflict, which has inspired many later attempts at marshalling technocratic principles in the hope of overcoming the perceived weaknesses and limitations of democracy (Maier 1970). There was no claim here of completing the democratic project; rather, the purpose was to finally overcome what Plato and many later thinkers saw as some of the principal weaknesses and dangers of democratic government. Arguments for technocracy can be made within a broadly democratic political order if they are cast as political projects in their own right. In  contrast, contemporary arguments leave us with the worst of both worlds. They function as attempts to replace the principle of popular sovereignty with one where decisions  about competences are taken by experts. And they do not  open up any challenge to democracy that could then be subjected to collective deliberation and public debate. This is not to suggest that the tension between technocracy and democracy can simply be resolved by asserting the primacy of the latter. Such an assertion does little to answer the concrete questions posed by complex policy problems that rely upon expert knowledge. The point, rather, is to argue

42  Christopher Bickerton et al. that the distinction between sovereign and government, and the primacy of popular sovereignty that it entails, can serve as a regulative ideal against which we can judge the balance between experts and political representative in any given case. This confronts more directly the issue of who has final authority, and does not  seek to deny the inherent tensions between claims of authority based on expert knowledge and those based on public authorization.

Conclusion As this chapter has shown, there has always been a deep tension between technocracy and democracy. In the classical case for technocracy, made by Plato, the goal is very clearly to replace democratic self-rule with a form of rule by experts. It would be inaccurate, however, to think of today’s technocrats as modern-day embodiments of Plato’s philosopher-kings, even if some of the contemporary discussion around the dangers of technocratic rule suggest this to be the case.5 In  fact, technocrats today, and the non-majoritarian institutions in which they work, are usually situated within a political landscape that is ostensibly attached to the principles of collective self-government and popular sovereignty. At issue, therefore, is not the replacement of democracy with technocracy, but the nature of the relationship between the two, and in particular the place of expertise within democratic political life. This chapter has argued that when framed in this light, contemporary arguments for technocracy are in fact much more of a challenge to democracy than they appear at first sight. Though a place is assured for electoral political competition, the common view is that technocracy is most suitable for certain kinds of policies. These arguments implicitly assume that the matching up of decision-making regime and policy issue is itself an academic exercise. In doing so, they take it upon themselves to decide what constitutes a policy area that needs to be subjected to public deliberation and political competition. At this final stage of the chapter, it is important to emphasize that in a political regime founded on the democratic principle of “identity” between the rulers and the ruled, there is of course a place for experts. Rousseau’s distinction between the “sovereign” and the “government” is once again useful here. As long as the sovereign power to make politically significant decisions remains unmistakably with the “people” as a whole, experts can function both as “advisory” bodies, providing relevant information before political decisions are made, and as “delegated” bodies, responsible for “enacting” these decisions after they are made by the people. From this point of view, the core of the contemporary technocratic challenge to democracy lies in the fact that it inverts the hierarchical relationship of authority between the sovereign and the government that is usually expected in a democratic system founded on popular rule. Instead, it places experts in the superordinate position of the sovereign and the “people” in the subordinate position of the government. In more recent versions of the argument for technocracy, this challenge is compounded by the fact that the transfer of sovereign power

Technocracy and political theory  43 from the “people” to experts happens under the false pretences of a purported “complementarity” between the two, which effectively undermines the scope for meaningful political contestation over the question of the relative merits of democracy and technocracy.

Notes 1 There  has been some debate about whether or not  European integration in the 2000s—with many instances of delegating power to expert bodies—was a case of a “constitutional moment”. This was the ambition of the Constitutional Convention presided over by the former French president, Valéry Giscard d’Estaing. The French and Dutch rejections of the so-called “constitutional treaty” in 2005 seemed to suggest that this was not a constitutional moment as such. What resulted was the Lisbon Treaty of 2008, ratified by national governments in the same manner as earlier EU treaties. The same can be said of the decision to create a common currency managed by an independent European central bank—an important decision, but not one that was absorbed into a wider moment of political revolution or change. It was simply decided upon by national governments in the context of existing EU procedures. 2 In making this argument, Monti and Goulard are relying on what economists call the “time inconsistency problem”. This  was first introduced into political economy in the late 1970s by Kydland and Prescott (1977) in a seminal article that explained why “policy rules” were preferable to “political discretion” in macroeconomic policymaking. The debate at the time was precisely around the difficulty for governments in containing inflationary pressures. The  understanding of inflation at the time, however, was not “technocratic”; rather, it was considered an outcome of underlying social conflict around distributional outcomes. See, for instance, the work of John Goldthorpe (1987). See also the discussion of “neoliberal technocracy” by SánchezCuenca in Chapter 2 of this volume, which discusses some of these themes. 3 Another effect of price stability is, perversely, to make it feasible for large businesses to hold off making capital investments. Higher rates of inflation can serve as an incentive to reduce to a minimum the cash reserves of a corporation, and one way of doing this is to invest in projects likely to enhance the long-term growth rates of a firm in a given industry. Today, some of the largest corporations in the world sit on huge cash piles, as they do not have the monetary incentives to draw them down that can exist in times of higher inflation. 4 See Bickerton (2011) for an extended critique of this “neo-Madisonian” argument, which extends beyond EU studies into scholarly work on international organization more generally. See for instance the work of Keohane et al. (2009). 5 The words of British politician Michael Gove, who, during the UK EU referendum campaign of 2016, argued that “the people have had enough of experts”, is one such instance. Others include responses to the introduction of technocratic rule in Greece and in Italy during the Eurozone crisis.

2 Neoliberal technocracy The challenge to democratic self-government Ignacio Sánchez-Cuenca

Introduction This chapter examines the tension between technocracy and democracy, making a crucial distinction between two conceptions of democracy: the individual or liberal, and the collective or aggregative. The individual conception sees democracy as an institutional device to limit political power and to protect individual freedom. The  collective conception emphasizes self-rule or self-government, which entails the involvement of the demos in the crucial choices about the kind of society that people would like to inhabit. The main argument of the chapter is that technocracy is a threat to the collective conception of democracy, but not necessarily to the individual one. By limiting the scope of democracy, technocracy mutilates popular sovereignty, which is at the basis of the collective conception. Democracy without sovereignty is but a method of selecting rulers, exactly in the sense that Joseph Schumpeter advanced in the 1940s: elections cannot be read as the manifestation of the popular will through the aggregation of preferences; rather, they are a mechanism of accountability by virtue of which the citizenry can get rid of bad rulers. Technocracy and individual democracy can coexist harmoniously, but at the cost of draining democracy of its transformative power. The  approach presented for the analysis of technocracy is different from, though not  necessarily incompatible with, Caramani’s (2017) analysis based on modes of representation. Rather than focusing on the representative link, this chapter examines technocracy from the broader perspective of theories of democracy, analysing in each case the role of the citizen in the political system. Technocracy stems from the tension between democracy and truth (see Figure I.1). If there are solid truths about the conduct of public affairs, why should decisions be left in the hands of the people? If there is a course of action that is objectively superior to others, what would be the point of electing politicians with different views on what should be done? The standard justifications offered for technocratic rule have some family resemblance. The underlying similarity comes from a sceptical feeling towards democracy. Some of the most frequent justifications are these: (1) experts and technocrats have longer-term horizons than politicians seeking re-election, (2) they are not subject to the pressure of special

Neoliberal technocracy  45 interests, (3) they know better than politicians and citizens, (4)  their decisions are evidence-based, and (5) their deliberations are of greater quality than those observed in representative politics. All these reasons are ultimately about the skills and knowledge of experts and they are meant to improve policy outcomes. Potential technocrats can be engineers, scientists and managers—that is, disinterested actors who might use their knowledge for the benefit of society. Technocracy may appear in very different political contexts (see the Introduction by Caramani). It can be found both in centrally planned economies and in neoliberal capitalism. In communist countries, scientists were supposed to solve the daunting task of how to establish the prices and quantities of thousands of goods, but also how to develop productive forces according to principles of scientific management (Hoffman and Laird 1985). In current neoliberalism, by contrast, technocrats are the watchmen of free markets. Free markets have to be protected from political interference. This  requires institutional means to stop democratic politics from undermining capitalism. The chapter focuses on the neoliberal variety of technocracy, which, arguably, is the most relevant and influential form today. Neoliberal technocracy shares many of the characteristics of traditional technocracy, but it also displays some specific traits that set it apart from the standard model. Neoliberal technocracy is intimately linked to the social and epistemological success of economics as the queen of the social sciences. Based on neoclassical economics and its variants, neoliberalism claims that pro-market economic theory should trump democratic decisions on the economy. This  can be achieved by approving constitutional rules that limit politicians’ discretion, or by delegating economic decisions to independent economic agencies that are “insulated” from the political game. Neoliberalism has an ambitious and aggressive agenda: it holds that an entire policy domain—that of the economy—should be largely removed from the democratic process. This  thesis assumes that market  allocation is intrinsically superior to the coercive methods used by political power. This  preference for depoliticization is justified not only in terms of efficiency gains, but also in terms of preserving and enhancing liberty. The constitutionalization of economic policy and the delegation of economic policy to technocrats imply a significant restriction on the scope of the democratic principle. It is important to emphasize here that neoliberalism does not justify itself by appealing to the superior technical capacity of economists, as traditional technocracy usually does. Neoliberalism is not exactly about a scientific management of society; rather, it seeks to impose a specific view about the role of markets and the state in the democratic process. Whereas the traditional technocrat has a broad and undefined mission, namely a more efficient management of social issues, the neoliberal technocrat has a more specific goal, the insulation of the economy from democratic politics. The neoliberal technocrats are not scientists or engineers with high problemsolving capacity, but economists or experts in finance who use their technical knowledge in service of certain political ideas about the relationship between democracy and the market. Neoliberal technocracy has a strong ideological

46  Ignacio Sánchez-Cuenca component that is absent in more traditional conceptions of technocracy, where technocratic government is conceived as the application of scientifically objective truths that overcome the ideological divisions that are an intrinsic part of democracy. The  chapter proceeds as follows. Firstly, it delineates two conceptions of democracy. Secondly, it identifies and analyses the tension between economics (as true knowledge about what is good for the economy) and democracy. Thirdly, it focuses on the key criticisms that neoliberalism has formulated against democracy. Finally, it analyses neoliberal technocracy from the viewpoint of the two conceptions of democracy.

Two views of democracy The extent to which neoliberal technocracy challenges democracy depends on the conception of democracy one holds. I will distinguish two conceptions: the liberal (individual) and the aggregative (collective). They differ with regard to what democracy produces. The liberal or individual conception considers that democracy is about leader selection. The voting procedure is used to fill representative positions. The aggregative or collective conception, by contrast, assumes that the ultimate goal of voting is to generate self-government or self-rule. Democracy is about making collective choices based on the people’s preferences. The  liberal conception of democracy is strongly influenced by Joseph Schumpeter’s classical work, Capitalism, Socialism and Democracy (1942). Schumpeter argued that the function of democracy is to produce a government through popular elections. In  his own words, “the democratic method is that institutional arrangement for arriving at political decisions in which individuals acquire the power to decide by means of a competitive struggle for the people’s vote” (1942: 260). Alternation in power is decided by the distribution of votes. Schumpeter’s view of democracy is so “minimalist” (Przeworski 1999) that he does not  even make room for the voters’ will. Given his scepticism about the judgement capacity of the voters, Schumpeter does not  allow any substantial connection between voters and leaders beyond the election process. The idea of a popular mandate, for instance, is devoid of any meaning in his theory. The connection between the people and their representatives is severed once the elections have taken place. From this point of view, elections are merely a mechanism of accountability (punishing the bad rulers and rewarding the good). Likewise, popular participation is not  a relevant concern for Schumpeter. Thus, a regime in which the vote is exercised by a minuscule proportion of the population counts as democratic as long as elections are held and there is competition for the vote. But this means that political equality is not a specific preoccupation in his conception of democracy. Other liberal theories have tried to expand Schumpeter’s theory in several ways. Dahl (1971), for instance, introduced participation as a key feature of democracy: in his theory of the polyarchy, the two basic dimensions of the form of government are inclusion (participation) and contestation (competition). Apart from high inclusion and contestation,

Neoliberal technocracy  47 polyarchy requires that certain conditions—such as freedom of expression, free and fair elections, freedom of information and others—are met. In the liberal conception, democracy is worthwhile for two reasons. Firstly, it is an institutional, non-violent way of replacing the rulers: in democratic states, rulers are changed without bloodshed (Popper 1945: 110). Elaborating on this idea, Przeworski (1991) argues that the democratic procedure avoids violence because elections generate an intertemporal calculation for political forces: losers comply with the results because they have a chance of winning in the future and this may compensate for the temporal cost of being in the opposition. Secondly, democracy is useful for limited government. Liberalism considers that the best way to secure liberty and the exercise of fundamental rights is to avoid an excessive concentration of political power. In a democracy power is more dispersed, among other reasons because rulers can be replaced periodically. Democracy, thus, is instrumental in limiting the exercise of power and, ultimately, protecting liberty (Saffon and Urbinati 2013). The liberal conception assumes that the value of democracy rests on outcomes that can be enjoyed by the individual. Once governments are formed through elections, individuals live in peace, they see themselves as politically equal, and they have liberty. These are outcomes that may be good for all, but they are not intrinsically collective in nature. Hence, the liberal conception can be deemed individualistic in its basic approach to democracy. The aggregative conception, by contrast, has an inherent collective dimension: it assumes that democracy hinges crucially upon the collective capacity of the people to rule themselves. The subject of democracy is a collective—the demos—whose members agree to make collective decisions together, aggregating beliefs and preferences through deliberation and voting. Political self-rule, as compared with the exercise of individual liberty, assumes a collective process of decision making that cannot succeed unless everyone is willing to participate in the aggregation process. The aggregative conception is able to capture some characteristics of democracy that are usually overlooked in the liberal conception, such as popular sovereignty and self-government. Popular sovereignty is embodied in the founding power of the people (Kalyvas 2005): in the constituent phase, the people have the capacity to create the institutional rules that define the political game. It is a pure act of political self-determination that generates a political order. Once the institutions are created, procedures are established to aggregate popular preferences into collective choices. Political institutions transform popular preferences into collective decisions. Self-government can simply be understood as collective decisions made according to the distribution of preferences in society. This is why democracy is traditionally defined as the rule of the people, in opposition to the rule of “the best” (aristocracy), the rule of the rich (plutocracy), the rule of the mighty (oligarchy), the rule of clerics (theocracy), the rule of dynasty (monarchy), or the rule of experts (technocracy). The aggregative conception does not necessarily require unanimity or the formation of Rousseau’s “general will”. The  will of all is sufficient. No organic

48  Ignacio Sánchez-Cuenca conception of the demos is needed either. In fact, majority rule can work as a privileged form of self-government. Even if majority rule falls short of unanimous opinion, public deliberation acts as its main legitimation instrument: people accept the taking of decisions by the majority if they have the capacity to participate in public debate on the alternatives (Manin 1987). The aggregative conception of democracy is also consistent with a notion of the common good, the general interest or equivalent expressions. This is obviously the case for the provision of public goods (many of which are valence issues in electoral terms), but even when there are deep divisions of interest, democracy may facilitate compromise and produce something close to the common good (Mackie 2009: 131). In distributional conflicts, compromise is indeed harder, but far from impossible. After votes are cast, revealing the popular support for the various alternatives, bargaining and interest accommodation can bring about a conciliation of opposing interests. The  capital–labour conflict was initially conceived as zero-sum, but the deep divisions were diluted in many countries thanks to class compromise (wage moderation in exchange for social protection) (Przeworski 1985). This holds even for moral conflict: despite the potential clash of absolutes, democracies can use a number of political devices to find common ground and intermediate solutions in matters such as abortion or euthanasia (Dworkin 1993).

Democracy and economic knowledge In the aggregative conception, democracy is about the people making collective decisions, normally through their representatives. However, if someone claims to have an objective, scientifically proven solution to public affairs, on what grounds could this be resisted? A perfect illustration of the conflict between democracy and truth is Henrik Ibsen’s play, An Enemy of the People (1882). In this story, the economy of a small town depends on the people who visit its baths, and a doctor finds out that the waters are contaminated. If the information is made public, the reputation of the place might be ruined. In a tumultuous town hall meeting, the people show their anger against the doctor; trying to protect their way of life by concealing the truth. The fact that Ibsen organizes the play around a health problem in which the doctor’s opinions should prevail over the people’s is quite revealing. The medical metaphor has a long pedigree; it goes back as far as Plato, who used it to ridicule democracy (first in Gorgias, and later in the Republic). Plato opposes cookery and medicine: while the former aims at pleasure, the latter aims at health; democratic politics is closer to cookery than to medicine. Public issues, related to the “health” of the polity, have to be handled by doctors or experts rather than by cooks. There is no reason to assume, according to Plato, that the people are the best judges of what is most beneficial for them as a political body (on Plato and technocracy, see Bickerton and Invernizzi Accetti in Chapter 1). Today, “political medicine” is embodied in economics. Economics provides the theoretical apparatus to calculate the consequences of policy decisions in

Neoliberal technocracy  49 terms of efficiency and distribution. It analyses how policies affect growth, unemployment, inflation, trade, and other economic variables. Ultimately, as a moral science, economics tries to find answers to the question of how to maximize collective welfare. Economics is by far the most successful, powerful and influential field in the social sciences. Its academic practitioners are better paid than colleagues in other fields (with the exception of engineers): they defend a compact, homogenous approach to knowledge, and they claim to have a more scientific methodology than other social scientists (Fourcade et al. 2015). Moreover, the economist profession is heavily involved in public and private decision making, from participation in corporations to positions of high-level authority in international organizations, regulatory agencies, public administrations and governments (Davies 2016, Markoff and Montecinos 1993). It  is unsurprising that the intellectual and institutional success of economics revives the underlying tension between democracy and technical knowledge. Politics today is very much about economic fundamentals. Governments are supposed to foster growth, bring down unemployment, tame inflation and keep the fiscal deficit under control. It is worth recalling how recent this development is. Until the 1950s the national account system simply did not exist in many countries, and therefore there was no common benchmark to evaluate economic outcomes. Earle, Moran and Ward-Perkins (2017: 15) show the exponential growth of references to “the economy” in British party manifestos starting from the mid-1950s to the present. Today, economic outcomes are one of the main pieces of information to judge a government’s performance, as the gigantic literature on economic voting attests. Since the economy and economics have gained greater prominence in contemporary politics, the potential conflict between democratic outcomes and the economist’s recipes increases dramatically. If the policies made by a representative government deviate from what economists recommend, and if economics can generate scientific knowledge about what is best for society, how can the democratic outcome be justified or legitimized? Should not the people refer all decisions concerning the economy to the economists? Of course, economists often disagree and have different policy views even if they share similar theoretical assumptions. But their internal disagreements are fewer than in other social sciences (sociology, political science, anthropology). Economic knowledge is more “monolithic” than others are. This facilitates the epistemic claim that economists know better.

The neoliberal mistrust of democracy It  is not  easy to pin down the exact nature of neoliberalism. The  relationship between neoliberalism and economic theory is a complex one. Of course, neoliberals speak in the name of economic science. Furthermore, some neoliberal tenets are expressed as direct results of economic theory. However, not  every economist is a neoliberal. It  might be said that neoliberalism is an economic ideology. It draws ideas from economics, but it goes beyond it (Przeworski 1992).

50  Ignacio Sánchez-Cuenca Several economic currents, not all of them sharing the same assumptions, are agreeable to neoliberalism, including Austrian economics, the Chicago School, the Virginia School of public choice, and German ordoliberalism. Jones (2012: 2) defines neoliberalism as “the free market ideology based on individual liberty and limited government that connected human freedom to the actions of the rational, self-interested actor in the competitive marketplace”. As a political philosophy, neoliberalism draws lessons from economic theory that support the superiority of market institutions over political ones in terms of efficiency and protection of liberty. As Mirowski (2009) has emphasized, this does not imply an anti-state position; the state needs to be redefined, since its main function is to secure the functioning of markets. Unlike classical liberalism, in which the market is able to assert itself, neoliberalism supposes that a state is necessary to keep the market free of political or democratic interferences. According to neoliberalism, the market should be at the centre of social life, as it avoids the many pitfalls of democratic politics. Rather than reviewing arguments about the virtuous side of markets, for the present argument it is more important to analyse the theses that are employed to devalue democracy. In this regard, the contribution of public choice to neoliberalism is a crucial one. Public choice can be broadly defined as the economic analysis of politics (Mueller 1997), but here I refer to a narrower development: that of the Virginia School under the intellectual leadership of James Buchanan and Gordon Tullock. The main idea of the Virginia School consists in problematizing the nature of the state and the government. In classical economic theory, the state is a benevolent agent, normally with a utilitarian inclination. According to public choice theorists, the state is a collection of institutions populated by self-interested actors who seek greater budgets, influence and power. Once the functioning of the state is analysed through the lens of neoclassical economics, all sorts of “government failures” may occur. Given the resources that lobbies direct towards influencing policy, public officials can fall prey to special interests, raising tariffs that limit international trade, introducing special taxes that distort incentives or approving regulations that hamper competition. The policy process, far from promoting the public interest, may bias public decision making in favour of privileged groups that act at the expense of society. In her study of the economic consequences of trade regulations in developing countries, Krueger (1974) used the expression “rent seeking”, which vividly captures the thrust of the public choice approach. The general idea is that government regulation generates rents that go to economic agents with privileged access to legislators. Regulation creates incentives for lobbying, and ultimately for bribery and corruption: the profits of firms depend not so much on their productive advantages, but on being favoured by public powers. This is also consistent with George Stigler’s (1971) theory of regulation: interest groups spend resources trying to influence the regulator, who ends up making rules in favour of regulated interests (this later became known as “regulatory capture”). Ultimately, the agenda of the public choice school consists in identifying government failures, the counterpoint to the market failures that Keynesians and

Neoliberal technocracy  51 socialists of all sorts have used as justifications for the state’s intervention in the economy (Buchanan 1979: 178). Democracy is in trouble because even if the common interest can be identified through majority rule or any other aggregative method, its pursuit fails because of the poor structure of incentives that decision makers face. The irony is that electoral incentives may make democracy derail: the policy process works in such a way that powerful minorities are able to curb the state in their favour, hurting the interests of the many. Particular interests are better organized than the general public: they have greater incentives to acquire detailed information and have definite preferences on policies. These interests prevail over the general will. Thus, the occurrence of elections can hardly be celebrated as a moment of triumph for the society at large, since representatives may be more responsive to special interests than to the majoritarian preferences, particularly in those areas in which rents are concentrated (for the benefit of producers) and costs are dispersed (for the harm of consumers). In the name of the very ideals embodied in democracy, the public choice theorist concludes that the democratic political process is counterproductive or inefficient: the public would be better off if the aggregation of preferences were done by the price formation mechanism of the market. A competitive market would prevent rent extraction and the abduction of the general interest by particularistic groups. From this perspective, markets generate outcomes that are better aligned with the voters’ will than those of the rotten world of politics. For  good reasons, the Virginia School figures  prominently in accounts of neoliberalism. However, in order to fully understand the neoliberal mistrust of democracy, another school, rarely mentioned in the literature on neoliberalism, has to be added—namely, the Rochester School of social choice. The fact that its main figures have been political scientists and not economists may explain its absence from the discussions on neoliberalism.1 Social choice is the analysis of collective decisions and the methods used to make these decisions. Its main result establishes that both the way in which alternatives are packaged and the procedures followed to count preferences affect the likelihood of the various potential outcomes. Given a distribution of preferences, different decision-making rules may generate different collective choices. This shows that the mapping of individual choices to collective choices is not neutral; it generates biases in favour of certain results. Thus, those who have the power to choose the procedures may tilt the balance in favour of their preferred alternative. The most important challenge to the idea of a well-behaved social choice function is Arrow’s Theorem (Arrow 1951), which establishes that no social choice rule can meet a set of reasonable conditions if there are at least three alternatives.2 The  Theorem establishes that if the social choice function is well constructed and the basic conditions are met, then the decision maker is a dictator. If the decision maker is a collective, the possibility of intransitive social choices (cycles) cannot be ruled out. Other results in social choice point in the same direction. The  possibility of manipulation through strategic voting, or the power of the agenda setter in

52  Ignacio Sánchez-Cuenca multidimensional politics to organize the sequence of decisions in such a way that her ideal point is finally the chosen one, reveal the obstacles that may emerge in the social process of preference aggregation.3 Arrow’s Theorem and subsequent developments have been the object of endless discussion. Here I am interested in the positions held by the Rochester School, founded by William Riker. An impressive number of authors have followed Riker’s (1982) influential interpretation, according to which Arrow’s Theorem implies that any democratic method of preference aggregation produces arbitrary results (Mackie 2003: 10–15). If collective choice cannot be read as a reflection of individual choices, there is no way to determine what society really wants. But if we cannot identify the true collective preference, then the very idea of self-government becomes hopelessly vague. If preferences cannot be aggregated in a meaningful way, the people cannot claim to rule themselves by voting. According to Riker, the collective or aggregative conception of democracy (what he calls “populism”) is doomed, while the individual or liberal conception is vindicated. Democracy works as a mechanism for power limitation. The people can get rid of the rulers if they think that they are not performing adequately, but that is all the potential that democracy possesses. The Rikerian interpretation of social choice is a more technical development of Schumpeter’s (1942) attack on the idea of the common good. After the crisis of welfare economics in the 1930s, Schumpeter assumed that interpersonal comparisons of utility are groundless, and concluded that the government could not maximize collective welfare.4 In his view, the inconsistency of utilitarianism precluded any association between representative government and the pursuit of the common good or the general will. As Riker would do a couple of decades later, he concluded that self-government is chimerical. Both the Virginia and the Rochester Schools provide abundant arguments for devaluing democratic politics. Democracy cannot produce self-government, and government failures are pervasive. The consequences are twofold: the aggregative conception of democracy is untenable, and democracy has to be kept at bay if market freedom is to prevail.

The depoliticized democracy of neoliberalism Neoliberalism does not reject democracy, but it forcefully defends a diminished form of it in which the scope of majority rule is severely constrained. In the neoliberal mindset, democracy is valued for its contribution to the preservation of freedom. More concretely, democracy is good because it generates alternation, weakening the politicians’ grip on the state. But, from this perspective, alternation is all that democracy should produce. The worst scenario for the neoliberal is an unfettered or unlimited democracy, in which the popular will rules and legislates on how the economy should be organized. In a famous interview given by Friedrich Hayek on his second visit to Pinochet’s Chile, the economist described democracy in these terms: “Democracy has a task that I call of ‘hygiene’, ensuring that political processes are conducted in a

Neoliberal technocracy  53 healthy way. It is not an end in itself. It is a procedural rule that has the objective of serving freedom. But in no way has it the same standing as liberty” (in Caldwell and Montes 2015: 301). In the most controversial part of the interview, Hayek openly admitted his preference for a limited (or liberal) dictatorship over an unlimited (or illiberal) democracy. Buchanan (1994: 8) said something very similar: “A democratically elected parliamentary majority imbued with socialist ideas and vision can destroy the potential value that may be forthcoming from an unfettered market economy as much or more than the activities of an authoritarian regime”. Hayek’s words and the context in which they were registered are of great interest for the argument of this chapter. Let me start with the context. Chile under Pinochet exerted a strong fascination for three of the founders of neoliberalism: Milton Friedman, James Buchanan and Hayek himself. All three visited the country: Friedman in 1975 (Klein 2007: 98−103), Hayek in 1977 and 1981 (Caldwell and Montes 2015), and Buchanan in 1980 (MacLean 2017: ch. 10). Chile was the first laboratory of neoliberal policies: the dictatorship banned unions, privatized social security and health care, introduced school vouchers and designed a new constitution that constrained the potential democratic evolution of the country. Moreover, the implementation of Friedman’s shock therapy to tame inflation had worked and the economy was booming after the “disastrous” socialist policies of Allende. Consequently, for neoliberal economists Pinochet represented an interesting experiment: one that showed the possibilities of a “liberal” dictatorship that fostered economic freedom. In the neoliberal scale of values, liberty reigns unchecked. Liberty finds its most favourable context in the competitive market, where the consumers’ preferences are impersonally aggregated and reflected in the price system, without any trace of coercion. In this regard, the institution of the market is vastly superior to democracy as an institution for processing information. Hence, if the choice is between a market economy without democracy and a planned economy with democracy, the neoliberal opts for the former. If the choice is between democracy with Allende’s socialist policies and dictatorship with Pinochet’s pro-market policies, the latter option is preferable for Hayek. The nationalization of firms, price control mechanisms, the protection of industry and other similar socialist measures are beyond the acceptable in the neoliberal ideology. The “hygienic” role that Hayek conceived for democracy becomes effective when democracy is reduced to a device for power limitation. This requires a drastic limitation in the scope of the democratic principle whereby self-government is removed from the set of ideals associated with democracy. Democracy is consistent with neoliberalism as long as it is just a peaceful method for selecting and replacing governments, with no capacity to distort the market order. The  most obvious method to tame democracy is to establish strong constitutional rules that reduce the discretion of assemblies and governments. This has been the favoured alternative of the public choice school. According to Buchanan, all the existing constitutions are too soft on democracy. No current political regime falls within his idea of a genuine liberal democracy, as they all

54  Ignacio Sánchez-Cuenca give too much leeway to the government in terms of financial, regulatory and fiscal capacity. In his own words, we need to address “the constitutional challenge of placing constraints on the authority of government over the operation of the economy” (Buchanan 1994: 8). Milton and Rose Friedman, in their best-seller Free to Choose (1980), ended the book with a catalogue of constitutional amendments that would entrench budget and debt caps, free international trade, ban controls over prices and wages, and establish tax limits and limits on money creation. This was “hardly an exhaustive list” (309): the goal was to have a Bill of Economic Rights that would constitutionalize economic policy. The underlying principle, as the authors said in their introduction, was this: “by enabling people to cooperate with one another without coercion or central direction, it reduces the area over which political power is exercised” (3). Hayek also believed that democracy had to be limited through constitutional devices. He  was more explicit about the institutional structure of a truly constitutional democracy. As he explained in the third volume of Law, Legislation and Liberty (1979: ch. 18), a system of government should be based on two representative assemblies and a constitutional court. The  first body would be the legislative chamber, whose task is to pass general laws that limit any form of coercive power. This chamber should be ruled by opinion, not interest. Thus, its members should not  be politicians seeking re-election, but rather outstanding and successful individuals, with a non-renewable mandate of 15 years. In turn, citizens would vote once in their lives, at the age of 45, and they could only vote for members of their same age. There would be elections on a yearly basis and every year 1/15 of the chamber would be renewed (the legislators would retire at 60). Since the assembly would be quite small, Hayek suggests that their members could be chosen indirectly: the 45-year-old people would choose some regional representatives who would select the members of the assembly. With this scheme, any possibility of a democratic mandate or of representatives being responsive to their electorate is diluted. Of course, ideology would play no role, since the people would choose representatives based on their merits and achievements, not  on their platforms. The  second assembly would be governmental. It would be similar to a parliament, with an “executive committee” chosen by the majority and subject to the control of the other political groups in the assembly. This assembly would play the role of a parliamentary government, whose executive power would be constrained by the general legislation passed by the legislative assembly. Hayek, and neoliberals more generally, understood the importance of a broad popular legitimation of the political system, but tried to downplay its political consequences. This tortuous relationship between neoliberalism and democracy is driven by the attempt to square two goals that in principle clash with each other: on the one hand, an institutional system that preserves economic freedom (the free rein of markets) regardless of people’s opinions on the matter; and, on the other, a state that looks after the good functioning of the market but that needs to be checked by the people to avoid an excessive concentration of power.

Neoliberal technocracy  55 The people have the right to control the rulers, but the people must be deprived of the possibility of telling the rulers what to do on economic issues. In other words, democracy is functional in the neoliberal world as long as economic policy is “depoliticized” (see Tortola in Chapter 3 in this volume). This is one of the defining features of the modern global economy according to Bourdieu (2002): the economic sphere is to be guarded against politics.

Neoliberal technocracy There is a significant difference between the standard and the neoliberal conceptions of technocracy. In the standard model of technocratic governance, policy is delegated to independent or unelected agencies whose legitimacy relies mainly on expertise. The experts who populate unelected agencies act independently of electoral or political calculations; they make decisions based on their technical knowledge on the subject matter. The  traditional line of defence for independent regulatory agencies is that they generate results that are Pareto optimal—or, to put it in another way, that regulation does not  produce winners and losers (Majone 1996). Consumer protection or environmental concerns are typically mentioned as examples of issues with potential efficiency gains that can be better achieved by agencies insulated from the political process. This is consistent with one of the defining assumptions of technocratic decision making—namely, that society is sufficiently homogeneous to identify general or common interests (Caramani 2017: 60). Technocrats are supposed to figure out the best means to satisfy these interests. They are not the enemy of the people; on the contrary, they work for the improvement of society as a whole. Neoliberal technocracy breaks significantly with this assumption and introduces a curious twist. In fact, the neoliberal doctrine does not necessarily rely on the notion of the general interest (though it often does). On the one hand, as was highlighted in previous sections, some trends within neoliberalism are quite sceptical about the definition of the general interest, which is perceived as a remnant of metaphysical thinking.5 On the other hand, the public choice school would question Majone’s distinction between efficiency and distributive considerations. Even if regulation is presented as a purely efficiency-enhancing activity, it is always possible to identify losers. In the case of consumer or environmental protection, the losers are the producers who have to bear the costs associated with higher standards. Neoliberals have a clear goal in mind that goes beyond the general interest: securing individual liberty through the promotion of competitive markets. In order to achieve this goal, they use two institutional devices: the constitutionalization of policymaking, and the delegation of policymaking to independent, unelected bodies. Both follow from the spirit embodied in the “rules rather than discretion” approach. Constitutionalization and delegation can be illustrated with the prominent example of monetary policy. One of the seminal contributions to the formation of the neoliberal zeitgeist was Milton Friedman’s monetarist theory, which was based on the assumption

56  Ignacio Sánchez-Cuenca that discretionary policy produces worse results than fixed rules. Friedman’s (1968) reaction against Keynesianism was based on the idea that inflation behaves better if politicians cannot tinker with it. According to the theory, monetary growth leads to higher inflation; while inflation may generate a reduction in unemployment in the short run, in the long run the effect is dissipated. This effect, despite its short-lived nature, is not innocuous: the unemployment rate eventually returns to its natural or equilibrium rate but inflation remains higher than before the government’s intervention. The reason is that economic agents learn to anticipate the government’s intentions, neutralizing the inflationary surprise effect of monetary expansion. Friedman famously recommended that monetary policy should be kept away from politicians’ hands: the economy would improve if the central bank followed an automatic rule of money creation (in the 3%–5% interval). Friedman’s ideas were refined and framed into general equilibrium theory by the introduction of rational expectations. The underlying principle is that even if there is a well-defined social objective function and policymakers have all the relevant information, the strategic reaction of those affected by the policy decisions may ruin optimality: “The reason that they [policymakers] should not have discretion is not that they are stupid or evil but, rather, that discretion implies selecting the decision which is best, given the current situation. Such behavior either results in consistent but suboptimal planning or in economic instability” (Kydland and Prescott 1977: 487). This is also the argument that is employed to defend the independence of the central bank (see, for instance, Barro and Gordon 1983). The government commits itself not to introduce inflation by leaving monetary decisions in the hands of an independent central bank, whose governor is not accountable to the executive or the legislative. The only mandate of the central bank’s governor is to avoid inflation. The  institutional solution consists in insulating monetary decisions from politics. Insofar as monetary policy is delegated to an expert, it is removed from the democratic domain. The constitutionalization and delegation of monetary policy are not exclusionary methods. Both can be combined in imaginative ways. For instance, in the German ordoliberal school, economic policy is supposed to be constitutionalized and the independent central bank plays a review role similar to that of a constitutional court. According to ordoliberalism, the separation of markets and politics is part of the economic constitution of a well-ordered capitalist economy. This economic constitution is mostly tacit, but neoliberals would like to make it explicit and formal.6 Analogous to the constitutional review of legislation practised by constitutional courts in liberal democracies, the tasks of an independent central bank are a sort of constitutional economic review of government’s and legislative’s economic policymaking (Laval and Dardot 2016: ch. 2). The instrument for this review is the monopolistic control of monetary decisions. Both constitutionalization and delegation are visible in today’s capitalism. Thus, a number of countries have introduced constitutional provisions that tie the government’s hands on economic matters, such as balanced budget and debt provisos. Balanced budget requirements are present in the constitutions of US

Neoliberal technocracy  57 states, though not at the federal level. In Europe, during the economic crisis initiated in 2008, several countries, under the pressure of the European Union and financial markets, amended their constitutions regarding the deficit and the debt (Germany in 2009, Austria and Spain in 2011, Italy in 2012, Slovenia in 2013). The most ambitious experiment in the constitutional entrenchment of economic policy is the European Monetary Union (EMU). Strict fiscal rules, accompanied by quasi-automatic sanctions, have been approved, reducing dramatically the competences of governments in macroeconomic policy. This  was clearly perceivable in the 1992 Maastricht Treaty, which set targets on public deficit, public debt and inflation; and it reached a new level in 2012, in the midst of the Euro debt crisis, when governments agreed on the Fiscal Compact, which requires, among other things, that all member countries pass balanced budget legislation (the so-called “golden rule”). Delegation to independent central banks has been even more successful than constitutionalization. Central bank independence has become pervasive in contemporary political systems, in both developed and developing countries (Crowe and Meade 2007). Governments have willingly renounced monetary policy. Taking into account that monetary decisions have spillover effects on fiscal policies and other economic variables, the degree of democratic control over the economy has shrunk considerably. Neoliberal technocracy also has a highly visible presence in international organizations such as the International Monetary Fund (IMF), a body populated by technocratic economists (Momani 2005) that for several decades has imposed structural adjustments inspired by the “Washington consensus” on many developing countries (and more recently, in Western European countries as well). The “Washington consensus” was a label created by John Williamson to summarize the ten main policy recommendations that were attached to financial aid, based on stabilization, privatization and liberalization. Williamson (1993: 1334) denied any relationship between the Washington consensus and neoliberalism, but the “market fundamentalism” of the IMF has led many to argue that there is such a connection (see Rodrik 2006: 974). But beyond the discussion on the neoliberal pedigree of the IMF, its legitimation mode is clearly technocratic, and wherever it has intervened, it has restrained, at varying degrees, the scope of democracy on economic issues. The combination of constitutionalization and delegation is best exemplified by the European Union (see Bickerton and Invernizzi Accetti in Chapter  1). In  particular, the EMU works on the basis of a strong constitutionalization of economic policy as well as the empowerment of unelected bodies such as the European Commission and the European Central Bank (ECB). The recent events of the Eurocrisis show that the traditional justification of technocratic arrangements in terms of efficiency gains does not hold. During the Eurocrisis, technocracy operated fully regardless of the efficiency/distribution distinction. The crisis, in fact, was marked by an acute conflict of interest between creditor and debtor countries. In this context, monetary decisions by the ECB could not  be neutral, even though they were taken by independent experts. Here, it

58  Ignacio Sánchez-Cuenca is crucial to underline the difference between independence and neutrality (Adolph 2013): independence implies insulation from representative powers and the citizenry, whereas neutrality requires not taking sides, something that can be impossible when decisions generate winners and losers. In the context of the Eurocrisis, a disproportionate part of the costs have been borne by the debtor countries (Copelovitch et al. 2016: 827). Regardless of whether this is considered a fair solution or not, the fact of the matter is that the decisions on the anticrisis policies were intrinsically political but were made by non-political actors such as the Commission and the ECB. From this perspective, the European Union is perhaps the most ambitious experiment in technocratic decision making (Corrias 2017, Sánchez-Cuenca 2017). Apart from the constitutionalization of economic policy, European technocratic agencies (mainly the ECB, but also the Commission, which is a curious mix of technocratic and political dimensions) play a crucial role regarding oversight of national budgetary issues, monetary decisions, banking regulation, competition policy and other related issues. Because of distributional effects, the legitimacy of these forms of technocracy is subject to controversy, as the copious literature on the democratic deficit of the EU attests. According to neoliberalism, economic technocracy is an institutional safeguard against the collectivistic temptation of democratic politics. It is not about improving democracy or democratic outcomes; rather, the assumption is that democracy is a dangerous political system for the preservation of liberty. Democracy has to be curbed, depoliticizing economic policy. The neoliberal justification of technocracy is quite different from the liberal one regarding limits on majority rule. According to classical liberalism, the fundamental rights of the individual cannot be trumped by majoritarian decisions. Division of powers, checks and balances, and constitutional review are mechanisms to control the abuses of the legislative assembly. By granting fundamental rights to every citizen, democracy is, so to say, protected from itself. In a liberal democracy, the majority cannot decide to alter the principle of political equality, which is at the very basis of democratic politics. Citizens cannot be excluded or marginalized from the democratic process even if a majority wants to do so. Here, the limitation of democracy is introduced for the sake of democracy. Neoliberalism is an altogether different approach. Democracy is constrained for the sake of economic freedom. The  people, or their representatives, cannot decide on economic issues—not because by doing so democracy would be undermined, but because the market economy would be impaired. Democracy cannot legislate on issues that, in the neoliberal view, each individual, following her tastes and wishes, should decide upon. To put it another way, economic freedoms are so highly regarded by neoliberals that they are located above the democratic process; they are as important as fundamental rights and deserve the same protection. Democracy, thus, is subordinated to a certain view of the market economy. This helps to explain the half-hearted commitment to democracy among some of the founders of neoliberalism that became obvious in the attitudes of Hayek or Buchanan towards Pinochet’s Chile.

Neoliberal technocracy  59 Neoliberal democracy goes farther than the liberal model in terms of restrictions on majority rule. It “amputates a vital organ”, such as economic policy, from the domain of democratic decision making. The bottom line is that it is not up to citizens to decide on economic issues. Under this approach, the citizenry, or their representatives, are powerless to interfere in the functioning of markets. The operation of the market is left in the hands of technocrats who make decisions on monetary, trade or competition policy; moreover, constitutional rules on the debt and the deficit prevent the people having any real influence on economic issues. As a consequence, democracy is chained by the imperatives of the market economy. The evaluation of the resulting form of government depends on what conception of democracy is held. Technocracy is largely consistent with what I have called the individual or liberal conception of democracy, in which the basic function of elections is to generate a government in a competitive way. Competition for power is intrinsically linked to the possibility of alternation. Alternation is crucial to limit power and, therefore, to protect individual liberty from potential state coercion. In this view of democracy, the scope of representative government does not  enter into the picture, and therefore the expansion of technocratic decision making is not necessarily a threat to democracy as long as elections take place and governments are formed. Alternation may happen, but a change in government cannot bring about deep changes in economic policy since the crucial decisions are either foreordained by rules or made by technocratic agencies. Nonetheless, the election of a government is still an important issue: the capacity and honesty of the rulers, and their decisions on foreign policy, moral values, infrastructures, education, social policy and many other fields, may be highly consequential for society. What is left out of the democratic game is the economic order of capitalism. The protection of competitive markets is the inner citadel of neoliberalism, its unassailable domain, beyond the reach of democratic politics. Technocracy becomes a more questionable institutional arrangement if the aggregative or collective conception of democracy is embraced. The  people’s capacity to make collective choices on what kind of society they want to live in is severely constrained by neoliberal technocracy. For the aggregative conception, the restriction of self-government impoverishes democracy. The  exercise of self-government, obviously, is intimately linked to the idea of sovereignty. Insofar as sovereignty might lead to the questioning of the economic order, it must be eliminated from politics. The most effective way of eviscerating sovereignty consists of undoing the demos—the collective that agrees to make political decisions in common (Brown 2015). In  both the liberal and the neoliberal versions of democracy, the demos is never theorized. Precisely because all collective elements are eliminated, the subject of the democratic game is irrelevant (democracy is reduced to a procedure, regardless of who participates in it). If there is no demos, there is no sovereignty. Neoliberalism supports democracy as an institution that generates individual freedom, but it rejects the possibility of the people using their individual freedom to join a collective project that could undermine the primacy of markets.

60  Ignacio Sánchez-Cuenca

Conclusion: sovereignty, technocracy and populism Using Fishman’s (2016) typology of democratic dimensions (authenticity, quality, depth and consolidation), neoliberalism eliminates the depth of democracy: without capacity for collective self-rule, democracy becomes flat. Democracy in neoliberal times can be a real democracy, and it can also be a well-functioning one, but it lacks capacity for economic change according to the people’s preferences. In  this light, Davies’ (2016: 6) critique of neoliberalism as “the pursuit of the disenchantment of politics by economics” makes sense: politics becomes disenchanted when its potential for transformation is eliminated. The curtailment of the democratic ideal of collective change may generate disaffection and despair. This opens up still another way for the understanding of the similarities and differences between technocracy and populism (see the Introduction to this volume). The two doctrines hold mirror-image views on sovereignty. The neoliberal technocrat despises the idea of sovereignty, whereas the populist embraces it with the utmost enthusiasm. As neoliberal constitutionalism and technocracy make inroads in representative democracies, a reaction of the opposite sign emerges under the form of new political actors that promise a restoration of lost sovereignty. The populist exploits the feeling of powerlessness shared by many citizens about the potential of democratic politics for collective change. Thus, contemporary populism can be seen as a political reaction to the neoliberal elimination of sovereignty from the democratic system.

Notes 1 In his history of neoliberalism, Jones (2012: 129) mentions Riker and the Rochester School as an extension of the public choice approach. 2 These conditions are the following: (1) the social choice rule must produce some collective decision; (2) the social choice rule must work with any kind of individual preferences (unrestricted domain); (3) if individual preferences are unanimous, the preferred alternative is the collective choice (Pareto principle); and (4) the choice between two alternatives is not affected by other alternatives (independence of irrelevant alternatives). 3 See Hinich and Munger (1997) for an overview. For  a discussion of the empirical relevance of these theoretical results, see Mackie (2003). 4 If interpersonal comparisons of utility are allowed, Arrow’s Theorem does not have implications for welfare economics, although it is still valid for voting procedures (Sen 2002: 77). 5 Any form of utilitarian reasoning is alien to Austrian neoliberalism. 6 As mentioned, Friedman talked about a Bill of Economic Rights that should complement the US Bill of Rights of the first amendment.

3 Technocracy and depoliticization Pier Domenico Tortola

Introduction “We have a Fed that’s doing political things … by keeping the interest rates at this level.  [T]he Fed is not  doing their  [sic] job. The  Fed is being more political than Secretary Clinton”. Thus spoke then-candidate Donald Trump (qtd in Blake 2016), during his first presidential debate with Hillary Clinton, expressing concerns about excessive central bank activism. Trump’s words echoed what Greece’s former finance minister, Yanis Varoufakis (qtd in Khan and Chan 2015), had said a few months earlier, from across the Atlantic and the political spectrum, about the European Central Bank (ECB): “[t]he attempt to ringfence Frankfurt from politics has produced a highly politicized central bank”. Statements such as these reflect a diffuse feeling—these days particularly common among populists—that unaccountable technocrats too often intrude into political matters which, in a democratic polity, ought to be left to the elected representatives of the people. And while central bankers—arguably the archtechnocrats—are probably the most frequent and visible target of such diatribes, the criticism applies equally to other areas. Consider, for example, Italian interior minister Matteo Salvini’s response to warnings by Tito Boeri (the economist chairing the country’s main pension fund, INPS—Istituto Nazionale della Previdenza Sociale), about the effect of harsh immigration policies and a planned pension reform on the sustainability of the country’s pension system: “[t]he president of INPS should worry about making the work of his agency more efficient … not do politics” (Meridiana Notizie 2018, author’s translation). But how does one square the foregoing with the view, expressed by, among others, pundit Tim Harford (2017), that one of the main problems of contemporary democratic governance is actually that “[t]echnocrats may not be too interested in politics, but politics is interested in technocrats”—or, put differently, that technocrats are more often the victims than the perpetrators of unwarranted turf invasions? Certainly, this will sound about right to anybody witnessing the more or less open attempts by the formally democratic Polish and Hungarian governments to tighten control over courts and independent agencies. Finally, and to add a further twist, how does either of the positions just described fit with the observation by Jean Meynaud (1969: 237), one of the earliest scholars

62  Pier Domenico Tortola of technocracy, that “the idea of an absolute separation between the technical element … and politics, is a myth … When he changes into a technocrat, the technician becomes a ‘politician’”? Partly, differences of opinion of the sort given above are due to interpretive disagreements on the available evidence, or to normative ones on the appropriate boundary between the technical and political sphere in public governance. However, the discord runs deeper than that, and stems from the coexistence of different conceptions of technocracy and of its relationship to politics, captured primarily by such terms as “depoliticization” and “politicization”. That these conceptual differences remain mostly implicit is unfortunate, for it fragments much of the scholarly and political debate into islands that do not communicate with one another as well as they should, and as a result are not as productive as they could be in answering important analytical and normative questions on the nature and role of technocrats in democracy. In an attempt to improve on the conceptual status quo, this chapter presents a systematic analysis of the technocracy–politics nexus that posits (de)politicization as the main dimension along which different definitions of technocracy vary. In doing so, therefore, the chapter attains a twofold goal: it clarifies the relationship between politics and technocracy, and it formulates a unified conceptual framework for the latter. The  next section begins the analysis by presenting a baseline definition of technocracy as a depoliticized method of decision making, located between full and no autonomy vis-à-vis majoritarian institutions. Movements along the autonomy spectrum politicize technocracy, with asymmetrical effects on the concept: as decision making loses autonomy, it becomes unequivocally less technocratic. As “technocracy-as-method” gains autonomy, however, it enters the territory of two distinct existing conceptualizations of technocracy—as a type of government, and as a type of informal regime—examined in eponymous sections. For each, the chapter looks at their relationship to politics, democracy and the remaining definitions of technocracy. Additional reflections are presented in the Section “Technocracy as a Type of Informal Regime” on the contested nature of “technocracy-as-regime”. The final section recaps and concludes.

Technocracy as a decision-making method A good way to start exploring a term “used as loosely as is technocracy” (Centeno 1993: 309) is to identify a simple, baseline version of it on which to build by logical steps. At its broadest level, technocracy is defined etymologically as rule by experts (see the Introduction to the volume). What is usually left implicit in this definition, but is no less important for the nature of technocracy, is that experts rule by expertise—i.e. by rationally applying their knowledge (acquired by study and/or practice) to a certain field. In its basic version, therefore, technocracy is a specific method of (public) decision making. Radaelli (1999a: 764) characterizes this technocratic “mode” as “based on the idea of the ‘one best way’ reachable by the ‘competent’ professionals who know the best means to an end”.1

Technocracy and depoliticization  63 How does this version of technocracy relate to the sphere of politics? Following Bernard Crick’s (1962: 16–17) celebrated reflections on the subject, politics can be defined as “the activity by which differing interests within a given unit of rule are conciliated by giving them a share in power in proportion to their importance to the welfare and the survival of the whole community”. Differently stated, politics is the process whereby different groups within a polity (institutionalized primarily, but not only, through parties) compete peacefully for the definition of society’s values, priorities and, ultimately, policies. It follows from the foregoing that the technocratic method of decision making is alternative to politics. Reaching a political decision entails a compromise among partisan preferences (however formed) consistent with the configuration of forces at play. Conversely, technocracy ignores, in principle, partisan positions and the existing balance of power to base its conclusions on the state of the art in the relevant area or discipline. A technocratic decision is, therefore, in essence depoliticized. Needless to say, these methodological distinctions tie into personal/professional differences between the politician and the technocrat: while the former has a more or less direct connection to a party and voters to whom he remains accountable, the technocrat is recruited based primarily on his sectoral expertise and intellectual abilities. In its purest form, technocracy as a method should be not only depoliticized, but also apolitical. While depoliticization indicates the shifting of decision making from the party-political to the expert arena, the notion of apolitical technocracy takes this process to its extremes. It depicts decision making as an activity reduced to mechanical rules and procedures, in which knowledge—in particular scientific knowledge—is the one and only factor at play, and is able to provide certain and univocal solutions to any policy problem. Apolitical decision making, Marcussen (2009: 377) notes, is “dehumanized, eliminating personal ideological and emotional features that escape calculation”. While conceivable in abstract terms, apolitical decision making is unlikely to ever materialize in practice, for at least two reasons. The first, epistemological, is that the production and the transmission of the knowledge on which technocracy rests are, at least to some extent, socially constructed activities reflecting the values and biases of researchers, both collective and individual (e.g. Feindt and Oels 2005, Fischer 1990, Jasanoff 1990). Although this is particularly true of the “soft sciences” such as economics, on which much technocracy is based (see Chapter 2 in this volume), the natural sciences are by no means immune from these dynamics. Few have articulated this view of knowledge as well as Michel Foucault (1980: 131): Truth isn’t the reward of the free spirits, the child of protracted solitude, nor the privilege of those who have succeeded in liberating themselves. Truth is a thing of this world: it is produced only by virtue of multiple forms of constraint. […] Each society has its regime of truth, its “general politics” of truth: that is, the types of discourse which it accepts and makes function as true; the mechanisms and instances which enables one to distinguish true and false statements, […] the status of those who are charged with saying what counts as true.

64  Pier Domenico Tortola Second, even when adopting a fully objectivist view of knowledge and science, the latter’s technocratic application to concrete cases will often require value judgements and choices among competing interests and preferences that cannot be reduced to expertise (Nelkin 1975, Turner 2008). In deciding whether to authorize a certain drug for the market, an agency will make decisions about safety thresholds and acceptable side effects; in setting monetary policy, a central bank will have to make calls about tolerable inflation and/or unemployment levels; and so on. Using Flyvbjerg’s (2002, 2004) formulation of the problem, the practical application of science—the transition from episteme to techne—usually calls, at least to some extent, for phronesis—namely, the handling of pragmatic and contextual value-based considerations and choices. Institutionally, technocracy as a depoliticized decision-making method requires being located, both de jure and de facto, somewhere in the middle region of the spectrum between full dependence and full independence vis-à-vis the majoritarian institutions through which politics operates (above all, parliaments and executives). On the one hand, technocrats need a non-trivial degree of autonomy in order to have decision-making room for manoeuvre. On the other hand, however, expert knowledge can plausibly guide public policy only if confined within the limits of some overall goal—in other words, a mandate—set politically. As Sartori (1987: 423) puts it, “[a] government of experts is admissible in regard to means, not ends”. So defined, technocracy is a familiar feature of modern democracies, embodied by institutions located at different points of the autonomy spectrum just described: from technical units within ministries and other executive agencies, which operate closer to political power, to highly autonomous bodies such as independent regulatory agencies and central banks. Courts could also be included in this baseline definition of technocracy, insofar as judges and prosecutors are considered technicians of the law, so to speak. Empirically, while countries use different mixes of technocratic decision making, a historical increase can be detected in the recourse to technocracy across Western democracies since the post–World War II years, and increasingly from the 1980s. This is particularly true for areas such as regulation, where the privatization and liberalization of a number of utilities sectors (combined, in EU member states, with the Europeanization of governance structures) have led to the multiplication of independent agencies (Gilardi 2005), and monetary policy, in which central bank autonomy established itself as the dominant institutional paradigm in pretty much all developed economies by the 1990s (Marcussen 2009, Roberts 2011). From a normative standpoint, the debate on technocracy-as-method remains wide open. On one side of the debate, supporters regard delegation of decision making to technocrats as not only necessary for tackling the increasing complexity of the business of governing advanced societies, but also desirable because it ties politicians’ hands in areas where short-term electoral incentives are detrimental to good policymaking, thus increasing the quality and credibility of public governance (e.g. Blinder 1997, Pettit 2004, Vibert 2007). On the other side of the fence, critics lament the lack of political pluralism, debate and accountability entailed by

Technocracy and depoliticization  65 the (fallible) technocratic method, and see the shifting of decision making to depoliticized arenas as little more than a way for governments to offload their political responsibilities (Burnham 2001, Flinders and Buller 2006, Roberts 2011).2 It is fair to say, however, that even among its critics, technocracy as a depoliticized decision-making method is not regarded as inherently and fundamentally at odds with the survival and thriving of a country as a democracy. In fact, besides probably being unfeasible, subjecting each and every stage and aspect of public governance to partisan political principles would likely cause more problems than it solves, not just in terms of policy performance but also with respect to the legitimacy of decision making. Removing some parts of policymaking from direct political control and placing it in the hands of experts can, in the last analysis, coexist with and even improve the quality of democracy to the extent that technocracy is circumscribed within its proper functional and normative sphere, corrected to mitigate its most negative features, and overall taken with a healthy grain of salt (Fischer 1990, Gilley 2017). The politicization of technocracy-as-method If methodological depoliticization necessitates technocrats to be located in a halfway position between full and no autonomy with regard to political actors and institutions, then the politicization (or repoliticization) of technocracy can be associated with movements away from that position. This is a logical conclusion which, as we shall see, is not always fully appreciated in the literature. The  most straightforward and broadly acknowledged mode of politicization is represented by losses of decision-making autonomy on the part of technocrats. As the political constraints on experts increase, the argument goes, their technocratic room for manoeuvre shrinks, and their decisions increasingly resemble those of mere executors of political instructions. Taken to the extreme, movements in this direction will eventually make technocracy lose its nature altogether and turn into administration in the narrowest sense of the term (on technocracy as “agent” see Chapter 12 and Introduction to the volume). Autonomy losses can take place formally as well as informally. Formal means include not only direct constraints imposed on the decision-making space of technocrats—for instance, through changing the statutory independence of a technocratic agency—but also, and perhaps more frequently, indirect measures that affect autonomy by acting on the technocratic body’s finances or its recruitment, promotion and dismissal rules (see Verhoest et al. 2004 for an overview). Taken together, these cases overlap to a great extent with the notion of politicization commonly employed in the literature on civil service (e.g. Peters and Pierre 2004, Rouban 2003). In addition to these formal constraints, technocrats can be politicized de facto whenever political actors manage to impose their influence on officially autonomous entities via personal, professional or ideological connections (e.g. Adolph 2013, Devins and Lewis 2008, Ennser-Jedenastik 2014, 2015).3 Politicization can also take place in the opposite scenario, namely if technocrats acquire so much autonomy as to escape political constraints altogether.

66  Pier Domenico Tortola In the absence of boundaries delimiting technocratic action, experts will necessarily have to act following policy ends set by themselves, whether openly or implicitly. But, as explained above, this by definition takes them outside of the technical realm and directly into the sphere of politics. Put otherwise, unlike the previous type of politicization, in which technocrats end up doing the bidding of some political actors, in this case experts become politicized by turning themselves into partisan actors working towards their own view of the “good society”. This form of politicization is every bit as important as the previous one. Yet it is often subtler, and less readily recognized by scholars and commentators alike. A  key reason for this lies in the asymmetrical effects of the two types of politicization on the concept of technocracy. While politicization as loss of autonomy is regarded as marking, at least in its extreme form, a departure from technocracy tout court—and hence a full semantic break with the term4 —politicization stemming from an excess of autonomy takes us outside of the confines of technocracy-as-method only to make us enter the territory of two additional conceptualizations of technocracy: as a type of government, and as a type of informal regime. These will be examined in the next two sections.

Technocracy as a type of government Politicization produced by excesses of technocratic autonomy can take place formally as well as informally. The  former case is the most straightforward, and hence will be discussed first. In its fullest form, the elevation of technocrats to formal political independence would take us outside of the confines of democracy and into those of authoritarianism. Perhaps the clearest example of such a situation is that of military juntas—to the extent, of course, that the military are included within the category of technocrats. More generally, any system in which the reins of power (or significant and protected portions of it, as for example in present-day Myanmar) are in the hands of people selected on the basis of their professional affiliation and competences, rather than through free and fair elections, has the traits of an authoritarian technocracy.5 Crick (1962: 105) describes this combination of technology and undemocratic politicization quite well in his discussion of decision making in communist systems: People in unfree societies which have abolished or forbidden the institutional means of making political decisions may think that the ideology—Marxian science, for instance—determines these allocations, so that everything remains simply a problem of application. But this, however much believed, is simply an error. […] What they are in fact doing, faced with all sorts of complicated choices and alternatives as to how to allocate scarce resources, is to make political decisions without the institutions and procedures which register actual social demands honestly. Because this chapter is mostly concerned with technocracy within formally democratic settings, we shall not elaborate further on the notion of technocratic

Technocracy and depoliticization  67 authoritarianism. There is at least another case, however, in which we can speak of a formal acquisition of political autonomy on the part of technocrats while remaining within the boundaries of democracy: this is the case of so-called technocratic governments (discussed also in Chapter 6). The latter expression usually indicates executives in parliamentary systems where the post of prime minister and at least the most important portfolios are held by experts with no party affiliation (although not necessarily without previous politico-administrative experience). The government headed by Mario Monti in Italy between 2011 and 2013 is probably the most prominent recent example of technocracy so defined. Relaxing the definition a little, the term “technocracy” is sometimes also used for mostly partisan governments headed by a non-partisan expert (such as the Papademos government in Greece in 2011–12) or, especially in presidential systems, for governments in which the key portfolios (above all the economic ones) are held by technocrats (Alexiadou and Gunaydin 2019, Dargent 2015, McDonnell and Valbruzzi 2014). Technocratic governments constitute the apex of political power for technocrats, who find themselves virtually unconstrained in their formulation and pursuit of policies, except for the same constitutional limits that would also apply to any party government. In fact, one might argue that technocratic governments are even less encumbered politically than party executives, as they are usually called in with a clear mandate to carry out deep social and economic reforms (on this, see also Valbruzzi’s analysis in Chapter 6). This is especially the case in difficult political junctures, such as those engendered by scandals and economic crises, in which traditional parties find themselves without the ability, willingness or credibility to carry out difficult measures (Brunclík and Parízek 2018, Wratil and Pastorella 2018). That such reforms are more often than not underpinned by a pro-market agenda only confirms the inevitably partisan nature of “technocracyas-government” (Alexiadou and Gunaydin 2019). At  the same time, as he is elevated to a ministerial position, the technocrat changes, at least temporarily, his formal status in the polity. While in a military dictatorship members of the junta are co-opted qua officers and remain in that role, the central banker who becomes finance minister in a democracy ceases to be a central banker and acquires a new role, inclusive of its recruitment mechanisms, for the duration of his tenure. This  is an obvious but important aspect, which has at least two key implications. The  first is that the relationship of technocracy-as-government with democracy is not as problematic as it is often depicted in political as well as academic discussions. Technocratic governments certainly weaken the representation and accountability links between government and voters. However, they act entirely within the same constitutional rules as any party government. One might argue, moreover, that technocratic governments are as much a symptom as they are a cause of the decline in the quality of democratic governance that is often associated with them, insofar as they fill a political vacuum left by ineffective party systems (see also Caramani in the Introduction to the volume). Their relationship with democracy seems, therefore, at worst ambivalent, but certainly not incompatible (Pastorella 2016, Schmidt 2011).

68  Pier Domenico Tortola The second implication is that the politicization of technocratic governments is largely in the open. While these executives sometimes employ a rhetoric of the necessity and inevitability of the policies they are introducing, for the most part the game they play is highly political, and it could hardly be otherwise given their institutional position. They are located within structures, procedures, and rituals that are eminently political: both the institutional ones, from receiving votes of confidence to finding parliamentary support for measures, and extra-institutional ones, such as being subject to the scrutiny of media and public opinion. They are entrusted with competences and responsibilities that would be hard to pass off as merely technical, and despite often resting on grand coalitions of parties, they do have an opposition that is there to remind them of the ultimately partisan nature of their choices. Commenting on the technocratic governments formed in Italy and Greece as a result of the Eurocrisis, The Economist (2011) remarked: Even a wholly technocratic government can never fully escape politics. In  any country powerful lobbies bargain and wrangle. In  a parliamentary system technocrats must deal with the partisanship and intrigues of an elected legislature (in Athens and Rome, lawmakers are eagerly waiting to trip up the newcomers). They also face public ire if they are seen as sharing out gains or pains unfairly. If technocracy-as-government transcends depoliticization not only in its methodological dimension (by acquiring political autonomy and therefore starting to determine the ends of public policy) but also in the institutional and discursive ones (by switching to a new formal status—that of executive power— and playing an openly political game), then what remains to connect this type of technocracy to the baseline conceptualization of technocracy-as-method? In other words, in what way, if any, can technocracy as a type of government be seen as depoliticized? The answer boils down to the personal/professional dimension: the only significant aspect of depoliticization that is left in technocratic governments is that its key posts are held by persons with a technical rather than a party background.6

Technocracy as a type of informal regime Technical experts do not necessarily have to be in government to have a say on the content and ends of public policies. They  can also acquire partial or total autonomy vis-à-vis majoritarian institutions while remaining formally subordinated to the latter. In this situation, technocrats are just as politicized as in the previous case, but their transformation into partisan players able to impose their preferences on public policy takes place in the shadows, so to speak, behind an official façade of rule by political parties in parliament and government. Rather than a specific type of executive, therefore, this version of technocracy is better described as a de facto regime, whose key feature is “[t]he administrative and political domination of a society by a state elite and allied institutions that seek

Technocracy and depoliticization  69 to impose a single, exclusive policy paradigm based on the application of instrumentally rational techniques” (Centeno 1993: 314). How do technocrats acquire such domination? Meynaud (1969) identifies four ways in which experts may “dispossess” politicians of their power. In  the first place, they can rule simply as a result of ample—nay, excessive—delegation of competences on the part of political authorities. This is especially the case when the terms of delegation are so vague as to render the expert “practically master of the problem” (Meynaud 1969: 80). Second, technocrats may disobey or resist, more or less overtly, the instructions and mandates set by politicians. Third, in their role as technical consultants and advisers to majoritarian institutions, experts can gain sway over political decisions by setting the terms of policy problems, pushing for their preferred solutions and downplaying or hiding disliked ones. (Needless to say, the more obscure the technical subject is to the politician, the greater the opportunities for this sort of technocratic domination.) Finally, technocrats may gain political influence by exploiting their role as mediators or coordinators of the work of other departments, particularly when inter-agency cooperation is difficult and the hard-won expert solution is unlikely to be questioned. Whatever the mechanism of dispossession, the latter radically alters the relationship between technocrats and politicians compared to what happens in the baseline conceptualization of technocracy as a decision-making method, despite the formal equivalence of the two situations. Technocracy-as-method is a tool in the hands of elected politicians, who use it to depoliticize part of the governance process for a variety of reasons, and more or less strategically, but always remaining in control of the perimeter of expert action. In technocracy-as-regime these roles are reversed, with technocrats being usurpers of the real power in the polity, and politicians the (more or less aware) victims. That the acquisition of political autonomy by technocracy-as-regime takes place in a covert fashion produces very different consequences compared to the case of technocracy-as-government. In  the first place, while the latter is, as explained above, at worst ambivalent vis-à-vis democracy, technocracy-as-regime constitutes an unquestionable violation of democratic principles, for it takes ultimate political power away from electorally legitimized institutions and places it in the hands of personnel without any (even indirect) popular mandate or accountability for their policy decisions. In technocratic regimes, democracy becomes to a great extent an empty ritual, which is renewed through elections every few years but is unlikely to be notably consequential when it comes to real power. The  foregoing is compounded by the second consequence of covert politicization—namely, that technocracy-as-regime justifies and reinforces its power by hiding its partisan nature. As they exercise political power, technocrats continue to deploy a depoliticized ideology and a discourse of expertise, impartiality, objectivity and pragmatism, and to present their policy solutions as the best and most rational, or even the only ones possible under existing circumstances (Caramani 2017, Putnam 1977).7 This fiction is made possible by technocrats’ formal position as servants of majoritarian institutions, which shields them from political exposure and responsibility. As Sarfatti Larson (1972: 5–6) puts it,

70  Pier Domenico Tortola the implicit appeal to an univocal notion of rationality, embodied in scientific and technical expertise, is inseparable from the depoliticization of social and economic issues. Thus, technocratic ideology legitimizes technocratic power by negating the political nature of the apparatus of domination in which technocracy is rooted (emphasis in original). That technocracy-as-regime remains depoliticized not  only at the personal/ professional level, but also at the institutional and discursive level makes this conceptualization definitionally closer to technocracy-as-method than is the case for technocracy-as-government (whose connection with the baseline definition is stretched quite thin, as shown above). Where the relationship between the three types of technocracy changes is in the normative sphere, in which technocracy-as-regime stands apart from the other two conceptualizations for being the only one unequivocally undemocratic. It is no surprise, then, that this variant of technocracy is the one more often giving rise to political as well as academic debates, as shall be discussed further in the remainder of the chapter. The contested nature of technocratic regimes Where can we observe technocracy defined as a de facto regime? Leaving aside the (far too many) extravagant accounts of the phenomenon that belong less in the realm of political analysis than in that of conspiracy theory, technocracy nonetheless remains quite a frequent occurrence. Studying 1960s France, for example, Meynaud (1969) claimed that an elite of grandes écoles graduates had appropriated significant chunks of policymaking power from their positions within the Grands Corps (such as the Inspection générale des finances, the Conseil d’état and Ponts et chaussées). Needless to say, that France was an archetypally strong state was both a cause and an effect of its technocrats’ vast power. More recently, the European Union has repeatedly been accused of being a system that allows unaccountable Commission officials and central bankers to rule over democratically elected member state governments (Radaelli 1999b). The EU, write Matthijs and Blyth (2017) “is a technocracy, run by bureaucrats who see rule-breaking as a challenge to their own authority”. On the other side of the Atlantic, the United States is not immune from domination by experts. Centeno (1993), for instance, sees aspects of technocracy-as-regime in such executive agencies as the Office of Management and Budget, with its key role in budgetary agenda setting. Expanding the analysis to the sphere of policy advice, Fischer (1990: 147) observes: It  is far from clear that policy experts are necessarily constrained by an agenda preestablished by those they serve. Indeed […] they can be deeply involved in determining the agenda. To be sure, such a role only confers an indirect form of power […] But it is nonetheless very real power that can be used at propitious moments to shape the course of events. Yet, for each of these examples it is possible to find opinions and analyses directed in the opposite direction. Writing in the same period as Meynaud, and acknowledging

Technocracy and depoliticization  71 the prominent place of the French grands commis among the country’s ruling elite, Ridley (1966: 38) nonetheless cautioned against concluding that real power had shifted away from party politicians to the technicians. Technocratic dominance over the state, he noted, “is generally proved by explaining why it must have happened, rather than by showing that it has in fact happened. The proof is theoretical and the conclusion not tested”. Balfour and David-Wilp (2016) provide a similar rebuttal with respect to the EU, whose “dictatorship of technocrats and a bloated bureaucracy” they call the number one myth about the Union. Finally, in her study of scientific advice in the areas of environmental policy and food and drug safety in the United States, Sheila Jasanoff (1990: 208) concludes that, despite the increasing role of expertise in federal decision making, “regulatory science continues to be debated and interpreted within the structural constraints of an administrative process that still is heavily dominated by politics and law”. Politically accountable officials in both the legislative and the executive branches, she continues, retain ultimate control of the advisory process, primarily through their power to decide whether and how to receive and accept policy advice. There are at least three reasons why the empirical application of technocracy defined as an informal regime is more contentious than is the case for the remaining two conceptualizations of technocracy. The  first (already touched on earlier) is that technocracy-as-regime is much more normatively loaded than both technocracy-as-method and technocracy-as-government. Saying that a state (or some other kind of polity) is a technocratic regime is saying that it is, at least to a significant extent, undemocratic. This makes technocracy a tool in rhetorical battles in which the term is used (or rejected) with respect to this or that political system (but also government, party or ideology) depending on one’s political position and interests. And while populists from left and right are especially fond of the word—for in their eyes technocrats are part and parcel of the elites working against the will of the people (e.g. Müller 2016)—they are by no means the only ones who have deployed the notion of technocracy for electoral gain. The second reason is conceptual: it has been argued here that the key defining feature of technocracy-as-regime is the experts’ de facto transcendence of the autonomy perimeter set by political authorities, which takes them from the realm of policy means to that of policy ends. Very often, however, tracing the line between ends and means is easier said than done. This is particularly the case for those expert bodies entrusted with broad mandates, which may encompass policy options so diverse in their characteristics and effects as to be themselves subject to being framed as policy ends. For example, a central bank instructed to preserve price stability within the economy not only has to define what price stability means in the first place, it might also have a range of choices in the pursuit its goal—e.g. whether and to what extent to purchase treasury bonds—that can in turn have important consequences on, among other things, the sustainability of public debt, the fiscal space of a state and, ultimately, the performance of an incumbent government. Finally, technocracy-as-regime is empirically contentious because of the many challenges presented by the empirical operationalization of this concept, even when policy means can be clearly told apart from ends at the theoretical level.

72  Pier Domenico Tortola A case in point is the role of experts as advisers of political actors and institutions. To what extent is a certain policy measure informed by scientific advice a manifestation of technocracy, with experts subtly manipulating politicians, or of a much healthier effort on the part of the latter to gain more information in order to make a better final decision? More often than not, the exchange process and the dynamics between the expert and political levels are too fluid to lend themselves to this sort of dissection. Another clear example is that of the resistance, boycott or violation of politically set mandates on the part of technicians, which can be very difficult to substantiate even when they are dealt with at the legal level—as, for instance, the recent court battles on the legitimacy of some of the European Central Bank’s crisis measures demonstrate.8 Taken together, all these aspects place technocracy-as-regime squarely into the category of “essentially contested concepts” (Connolly 1993, Gallie 1956). It is a paradox of these concepts that they are contested largely due to their social and political importance, but it is precisely because of their importance that scholars should strive to hone and clarify them as much as possible—ultimately making them less contentious. Technocracy-as-regime is a tricky case in this respect: grey areas, both theoretical and empirical, may never be eradicated completely. Nonetheless, noteworthy advances have been made, over time, in trying to trace the contours of this phenomenon more neatly (e.g. Radaelli 1999a, 1999b, Tortola 2019, Tortola and Pansardi 2019). Within the broader research agenda on technocracy this is certainly one of the most challenging tracks, but one that should nonetheless be accorded high priority for the analytical and normative payoff at stake.

Conclusion This chapter has achieved two related goals: first, it has systematically examined the relationship between technocracy and politics, which is primarily expressed by the notions of depoliticization and politicization. In doing so, the chapter has also expounded a unified conceptual framework for technocracy, which contains the three main usages of the latter term: technocracy as a decision-making method, as a type of government, and as a type of regime. Table 3.1 summarizes this framework by listing the main characteristics of each of the three conceptualizations of technocracy. In ordering the principal ways in which technocracy is used in existing political, journalistic and scholarly debates, this analysis has clarified the concept in an inclusive fashion. In other words, the goal here has not been to privilege one “truer” definition of technocracy over the others, but rather to shed light on the varied connotations of this important term and on the links between its several facets. Technocracy is a broad concept that is used in different ways by different people, at different times and for different purposes. To it we can apply much of what Hanna Pitkin (1967: 10–11) once said of the concept of representation: We may think of the concept as a rather complicated, convoluted, threedimensional structure in the middle of a dark enclosure. Political theorists

Technocracy and depoliticization  73 Table 3.1  Technocracy and depoliticization: a unified conceptual framework Technocracy as a…

Dimensions of depoliticization

Decision-making method

Type of informal regime

• Personal/ professional • Institutional/ discursive • Methodological Means

• Personal/professional • Personal/ professional • Institutional/ discursive

Main policy object Relationship with Compatible democracy • Independent Examples regulatory agencies • Independent central banks • Ministries’ technical units

Type of government

Ends and means

Ends

Distortive

Compatible to ambivalent • Dini government (Italy, 1995–96) • Fischer government (Czech Republic 2009–10) • Papademos government (Greece, 2011–12) • Monti government (Italy, 2011–13)

(All contested) • France (e.g. grand corps) • European Union (e.g. Commission, ECB, independent agencies) • United States (e.g. policy advisers, federal agencies)

give us, as it were, flashbulb photographs of the structure taken from different angles. But each proceeds to treat his partial view as the complete structure. It is no wonder, then, that various photographs do not coincide, that the theorists’ extrapolations from these pictures are in conflict. Yet there is something there, in the middle in the dark, which all of them are photographing; and the different photographs together can be used to reconstruct it in complete detail. That there is a complicated structure there in the middle means that while we can and should embrace conceptual pluralism, we should also strive to find common linguistic ground to allow various users of the term “technocracy” to understand and dialogue with one another. This  is what the foregoing analysis has tried to achieve. To go back to some of the questions raised at the beginning of the chapter, technocrats in a democracy can be both victims and perpetrators of politicization, depending on whether they lose or gain too much autonomy with respect to political authorities. Perhaps more importantly, this chapter has shown that technocracy can be political and depoliticized at the same time and in different ways, based on the dimensions of (de)politicization we are looking at. Accordingly, modes of politicization will change in different cases, and so will our normative assessment of it.

74  Pier Domenico Tortola For example, for a technocrat abiding by the boundaries with politics, the space for politicization is above all methodological: will he keep following his expertise, or will he start making decisions based on (undesirable) political criteria? However, in technocracy-as-regime the only way left is what Frank Fischer (1991: 340) refers to as “politicizing the political”: namely, acknowledging the inherently partisan nature of experts’ actions—something which might not improve things much, but which would at least increase the transparency of the political status quo. Political disagreements over the proper role of technocracy and politics in public governance certainly do not end here, and will likely continue for as long as democracy exists. What the reflections presented in this chapter can do, however, is to demarcate more sharply the analytical field on which these battles are fought.

Notes 1 The  reader should note that other authors in the volume (see e.g. Caramani in the Introduction) may use the term “technical” rather than “technocratic” to indicate the expertise-based method of decision-making described here. 2 An attempt to transcend the traditional dualism between responsibility and responsiveness, on which much of this normative debate is based, is presented by van der Veer in Chapter 4 in this volume. 3 Technocratic bodies can, of course, be influenced or even captured by societal actors that are not strictly speaking political—for an example, see Braun (2018) and Jacobs and King (2016) on the sway of the financial sector over, respectively, the European Central Bank and the Federal Reserve. Insofar as these societal actors are seen as carriers of partisan interests, these cases are analogous to the type of politicization just described. 4 This is the position taken in this book overall. See the Introduction to the volume. 5 Unlike Caramani in the Introduction to this volume, I refrain from using the term “regimes” to indicate these cases of technocratic authoritarianism in order to avoid confusion with the different meaning of “technocratic regimes” proposed in the next section of this chapter. There the noun “regime” is used to signify informal patterns of power and influence within a formally democratic framework. 6 In this respect, it might not be a coincidence that Italians—who have gained some experience in this area over recent decades—do not refer to this type of executive as technocratic government, but rather “government of technicians” (governo tecnico or governo dei tecnici). 7 This discursive depoliticization is not too different from the notion of depoliticization as denial of political conflict discussed by Caramani in the Introduction to the volume. 8 See, most notably, rulings by the Court of Justice of the European Union in Peter Gauweiler and Others v Deutscher Bundestag (case C‑62/14, judgment of 16 June 2015), and Weiss and Others (case C-493/17, judgement of 11 December 2018). By establishing the consistency of, respectively, the European Central Bank’s Outright Monetary Transactions (OMT) and quantitative easing (QE) with the Bank’s legal mandate, the Court essentially rejected the plaintiffs’ interpretation of the ECB as a technocratic usurper of power within the Monetary Union.

4 Technocratic responsiveness Reinout van der Veer

Introduction On 29 December  2017, US President Donald Trump tweeted that New Year’s Eve 2017, which was going to be the coldest one on US records, could use a little global warming. The tweet was widely interpreted as an attack on the academic consensus on man-made climate change and sparked reactions from a variety of leading climate scientists. True to their nature, these climate scientists pointed to the logical flaws in Trump’s reasoning and provided evidence of the longterm rise in global temperatures. In the area of climate science, this example is one among many: scientists have frequently responded to sceptics of man-made climate change, and have even published articles on how to best debunk climate change deniers (Diethelm and McKee 2009, Van Rensburg and Head 2017). What is remarkable about this example is not  that there still are politicians who are avid climate change deniers and who are hostile to climate scientists, but that scientists decided to respond to their provocations. By engaging deniers on a public forum such as Twitter, scientists are in essence issuing a response to a political challenge. While such responses characterize the social media era that we live in, they also sit uneasily with the popular conception of the scientist as an “objective” actor that provides scientific input to an exogenous political debate. In a similar vein to President Trump’s frequent assaults on the field of climate science, populists have recently attacked technocratic elements in liberal democracies for being unresponsive to the “will of the people” (Mounk 2018). Efforts to depoliticize issues through the delegation of political power to independent, expertise-driven actors in liberal democracies have created political systems in which there is increasing disconnection between what “ordinary citizens” want and what the “elitist state” does—or so the argument goes (see also Chapter 3 in this volume). Indeed, technocrats are commonly assumed to be unresponsive to political developments in their environment (Caramani 2017, Eggertsson and Le Borgne 2010, Esmark 2017). They are brought in to provide effective, efficient and stable solutions to complex problems that ordinary politicians are unfit to tackle, either because the latter are subject to electoral incentives or because they lack the necessary expertise (see Introduction to this volume).

76  Reinout van der Veer As ideal-typical technocrats are guided by science and expertise rather than democratic will and are not pursuing re-election, they should not be concerned with political threats to their functioning. Hence, they are assumed to be unresponsive. Yet if experts, including climate scientists, are responsive to political challenges when they do not hold public office, why should this differ when they do occupy such positions? Moreover, if “ordinary” bureaucrats are responsive to political threats (e.g. Alon-barkat and Gilad 2016, Gilad et al. 2013), why would the obtainment of a Ph.D. change that behaviour? This chapter challenges the conception of technocracy as an insulated, unresponsive form of government, and asks under which conditions and in which contexts technocracy is responsive to political developments in its broader environment. It is mainly concerned with technocratic executives, i.e. executive actors that enjoy at least some degree of independence, giving them discretion over political outcomes, and that legitimate their decisions and activities by means of science and expertise.1 The following sections draw on the literatures on political responsiveness, political reputation management and political signalling to construct a theory of “technocratic responsiveness” and formulate expectations on how such technocratic responsiveness may operate empirically. The subsequent survey of the empirical literature provides illustrative examples of technocratic responsiveness in the area of European Union (EU) governance. The concluding sections discuss avenues for further research and the normative implications of technocratic responsiveness for the technocratic challenge to democracy.

A critique of the responsiveness vs responsibility dichotomy The principle that holds that democratic governments need to be both responsive to the demands of voters and responsible in making policy has been central to political science theories in the post-war period (e.g. Pennock 1952; see also Introduction and Chapter 1 of this volume), and can be found in disciplinedefining works such as Scharpf’s (1999) concepts of input and output legitimacy. However, it was not until the work of Mair (2009) that the distinction was explicitly theorized. In the context of democratic political systems, Mair explained how political parties face a tension between optimizing the dimension of responsibility while at the same time remaining responsive to citizens whose preferences may diverge from responsible action. As a consequence of this tension, parties “might well be able to represent without governing, but they have difficulty when they seek to govern without representing” (Mair 2009: 10). Other scholars have built on Mair’s responsiveness–responsibility dichotomy. Linde and Peters (2018), for example, have argued that being responsive to voter demands helps parties build the necessary political capital, or “goodwill”, to be responsible in the future. Goetz (2014) has discussed the temporal dimension of the dichotomy, arguing that responsiveness is usually short-term, while responsibility is long-term oriented.2 Yet the tension between responsiveness and responsibility has recently garnered more attention in the broad debate on the state of liberal democracy and the rise of populism in Western societies (Caramani 2017,

Technocratic responsiveness  77 Mounk 2018, Mudde 2004). These authors attribute a strong desire for responsiveness on the part of the government to the populist ideology, which stresses the antagonistic relationship between the “pure people” and the “corrupt elite”. The populist conceptualization of responsiveness is a rather simplistic one: it implies that governing actors should straightforwardly follow the “putative will of the people” (Caramani 2017: 62). In theory, this will is unitary and objectively discernible, yet in practice it is often determined, or interpreted, by populist leaders themselves. Caramani (2017: 62) rightfully observes that this will is particularly putative when it is “extrapolated from a majoritarian or plebiscitarian vote whereby the majority is equated with society as a whole”. In practice, polling only occurs once every few years, and the congruence between polling results and the actual distribution of political views in society fades well before new elections are held. Moreover, populists are becoming increasingly gifted at delegitimizing polling outcomes, which allows them to construe the “will of the people” even if polling outcomes are not in their favour. In practice, all that is required for populists to claim responsiveness, and by extension unresponsiveness on part of their opponents, is to politically construe the support of an ad hoc simple majority and to equate this simple majority with society as a whole.3 In  contrast, a predisposition towards the element of responsibility is attributed to liberal and, more importantly, technocratic ideologies (Caramani 2017, Mounk 2018; see also the Introduction to this volume). Technocracy, with its reliance on science, expertise and “objectivity”, should provide stable, effective and non-partisan solutions in the face of complex and politically contested societal problems (Majone 1994, Putnam 1977, Radaelli 1999b). Technocratic actors are brought in when legislators seek credible commitment to long-term policy objectives, as delegation to independent technocratic executives helps legislators to ensure policy stability (Franchino 2002, Majone 2001). By placing legislators at arm’s length, delegation prevents current and future legislatures from defaulting on long-term commitments when incentivized to pursue short-term gains. Moreover, ideal-typical technocracy also rejects political input from below. Like populism, technocratic ideology presumes a non-pluralistic view of society in which there is a unitary common good (Caramani 2017, Introduction to this volume). However, unlike populism, technocrats believe that citizens are uninformed and uninterested, and thus believe that following the “will of the people” is irrational.4 The  heavy skewedness of technocracy towards the responsibility side of Mair’s (2009) dichotomy has rightfully led a range of scholars to warn of technocracy’s undemocratic qualities. For example, Caramani (2017) has described how populism and technocracy both present similar challenges to party-based representative democracies, although they differ on key aspects such as how they establish their authority. In a similar vein, Mounk (2018) describes how populists turn liberal democracies into “illiberal democracies” on the one hand, whereas technocrats push liberal democracies towards “undemocratic liberalism” on the other (see also Chapter 2, this volume). Under the populist conceptualization of responsiveness, it is indeed hard to imagine how ideal-typical technocratic actors would be responsive to the will of the people.

78  Reinout van der Veer However, a more accurate picture emerges if the simplicity of the populist conceptualization of responsiveness is problematized. Societies consist of multiple, dynamic coalitions of stakeholders and audiences whose preferences for government behaviour are heterogeneous and constantly evolving. Moreover, while some actors demand responsiveness, some also explicitly demand responsibility. This heterogeneity of preferences implies that when executives are faced with pressing issues that demand action, their responsiveness will always be welcomed by some actors and audiences but rejected by others. The evolving nature of these preferences also implies that popular, and thus responsive, decisions taken at a given moment (e.g. free healthcare through deficit spending) may later on turn out to be unpopular (e.g. high healthcare costs and an unsustainable government debt), or vice versa. To meet such complex and ever-changing societal demands, executive actors need to balance being responsive to multiple current and future stakeholders at once. Consider a classic and popular example of executive technocracy: the independent central bank.5 In line with the raison d’être of all technocratic executives, monetary policy was delegated to independent, expertise-driven central banks because it makes monetary policy more credible (Majone 2001). Monetary policy is placed outside the sphere of direct influence of elected politicians because the latter face short-term electoral incentives that harm an economy’s stability and growth in the long term.6 And because the success of monetary policy strongly depends on the confidence of market actors in the responsibility of the monetary executive, delegation of this task to an independent, responsible and competent central bank ensures this credibility. Clearly, independent central banks are not responsive to the people in a populist sense. They are expressly made independent in order to achieve efficiency gains, but also to prevent politicians from using monetary policy to “pander to the people” by using the monetary printing press for popular yet irresponsible policy (see also Chapters  1 and 2 of this volume). As Mair (2009) noted, the demands of the populist type of responsiveness are to voters and to parliament, but the demands of responsibility are to a much broader set of actors and stakeholders. This latter group includes actors in domestic and international markets, such as traders and investors (including many citizens of the state on whose behalf the central bank operates), who have a preference for stable and growthoriented monetary policy. Moreover, even if one is able to empirically capture or construct the “will of the people”, and if that this will is then found to favour irresponsible monetary policy, responsiveness to this will still implies unresponsiveness to the will of others. This is because if the central bank is to pander to the people through irresponsible monetary policy, it would likely also imply that it is being unresponsive to the will of the future legislatures and electorates who would pay the price for such short-term-oriented action. However, when democratic governments make “credible commitments” to the will of future electorates, they create constraints for current governments, which may end up at increasing odds with the demands of current electorates.

Technocratic responsiveness  79 From this example, it is apparent that the question of whether technocracy can be responsive cannot be answered by merely examining whether the behaviour of technocratic executives is congruent with the preferences of a simple majority of citizens at any given moment. Instead, the mechanisms through which technocratic executives engage with their political environments and the heterogeneous sets of audiences and stakeholders within them must be examined more closely.

A theory of technocratic responsiveness The functions that technocratic executives perform are vital to the functioning of democratic societies, whether these are the administration of a currency or the enactment and enforcement of complex regulatory regimes. Many of the activities of these executives revolve around the management, or regulation, of societal risk, i.e. threats to society.7 In principle, societal risk is the core reason behind the existence of government, whether it entails the risk of crime, invasion by foreign powers, flooding, disease or economic stagnation, to name a few. Yet theories of regulatory risk management explain how the management of societal risk creates another type of risk for such regulators: institutional risk. Institutional risks are threats to the regulator’s institutional position (Rothstein 2006, Rothstein et al. 2006). Societal risks rarely map onto the regulatory frameworks that regulators develop for them, and since the impact of regulators is often dependent on complex institutional regimes and requires time to be observable, observable success can be difficult to achieve. Regulators themselves have only a limited impact on the management of societal risks, and “the difficulty in satisfying conflicting demands on regulation, therefore, creates institutional risks that can threaten the legitimacy of regulatory organisations and their practices” (Rothstein et al. 2006: 95). An executive tasked with regulating the chemicals industry, for example, must accurately be able to assess the danger of certain chemicals to public health and the environment. If it fails to correctly assess the societal risk posed by a particular chemical, it may unnecessarily ban the use of that chemical and cause businesses to move to a country where regulation is more favourable. It may also be too lenient, however, causing the continued use of the chemical by industry to harm citizens and the environment they live in. Both over- and under-regulation of the societal risk posed by a chemical may in turn pose the institutional risk of regulatory failure and an associated loss of regulatory credibility. Regulators respond to these various types of risk by shifting their focus and action towards the type of risk they perceive as most threatening. Regulators who are concerned with societal risk allocate attention and resources accordingly—for example, by weighing the estimated impacts of these societal risks against the likelihood that they will occur (Black and Baldwin 2012). Other regulators may be more concerned with institutional risk. A  failure on the part of the regulator of the chemicals industry may lead to dissatisfaction amongst its key stakeholders, such as citizens, environmental groups or chemicals producers. Such a loss of legitimacy may in turn further weaken a

80  Reinout van der Veer regulator’s institutional position, compounding the institutional risk faced by the regulator. Risk-sensitive institutional regulators will therefore prioritize the mitigation of such institutional risks over societal risks when they prioritize tasks and allocate resources (Gilad 2012). Importantly, this sensitivity to institutional risk is not  unique to regulatory actors. All executive actors face some degree of institutional risk. Executives in general are competence- and budget-seeking (Leloup and Moreland 1978, Wildavsky 1964): they seek to expand their toolkits and budgets in order to be able to cope with new challenges and an ever-changing institutional environment. At a minimum, the preservation of competences and funds currently allocated to executives helps them ensure their institutional survival. Thus, any development faced by the executive, whether endogenous (e.g. regulatory failure) or exogenous (e.g. political opposition), that may reduce its budget or (partially) remove its competences poses an institutional risk to the executive. Moreover, like regulators, these executives adapt their behaviour in order to negate risks and bolster their position. In  most situations, however, it is perceived and not  actual competence that shapes an executive’s institutional position. An executive’s reputation is a vital asset in safeguarding its institutional position (Busuioc and Lodge 2016, Carpenter 2010, Carpenter and Krause 2012). Stronger reputations lead to greater discretion and autonomy, whereas weak reputations may result in repealed competences or budget cuts. Such reputations are constructed in the interplay between executives and the stakeholders and audiences in their institutional environment. Hence, it is the audience’s positive judgement of the efficacy, expertise and legitimacy of the executive that secures the executive’s institutional survival. This  means that executives that are sensitive to institutional risk are by extension also sensitive to reputational threats, i.e. threats to an executive’s reputation. The quest for positive evaluation of their activities by key audiences makes executives sensitive to a variety of reputational threats, such as public allegations of regulatory failure (Maor, Gilad and Bloom 2013), public opposition (Alon-barkat and Gilad 2016, Jennings 2009), political intervention (Gilad 2015) and media attention (Maor and Sulitzeanu-kenan 2016). Responsiveness to different reputational threats is guided by the executive’s institutionalized organizational reputation and identity (Gilad 2015). Ultimately, an executive’s legitimacy is a product of this selective responsiveness to reputational threats (Rimkutė 2018). The  consequence of the importance of institutional risk and reputation for executive survival is that executives are both the senders and the recipients of political signals. Theories of political signalling explain how executives are “constantly monitoring the political pulse for any signs of activity that may intrude on ‘business as usual’” (Worsham and Gatrell 2005: 367; see also Bailey et al. 2005, Huang 2015). In response to such signals of institutional risk, executives will actively signal key audiences in this environment to preserve their reputation. What constitutes such a signal or a key audience is largely dependent on the context in which the executive operates, but the generic implication is that the

Technocratic responsiveness  81 executive will visibly alter its communication with or behaviour towards its environment in order to appease what it sees as audiences that are constitutive to its institutional survival. Moreover, while executive signalling may be merely symbolic or have tangible consequences, it is never risk-free for the sender. Signalling intent without following through results in a loss of credibility among observing audiences. Such audience costs therefore tie signalling actors to a certain course of action (Fearon 1994, 1997) and increase when the behaviour of the executive is more salient and contested, or politicized.8 Thus, signalling is not necessarily cheap-talk, but it often precedes concrete action. Returning specifically to technocratic executives, the concepts of institutional risk, reputational threats and executive signalling are very useful, at least analytically, in problematizing the assumptions behind the unresponsive technocrat. Like any other executive, technocratic executives face institutional risk. Moreover, technocratic executives are solicited when there is a need for competence, expertise and objectivity. It  is therefore also these elements that are constitutive to a technocratic executive’s reputation, institutional identity and, ultimately, its survival. When technocratic executives receive signals that point to institutional risks and reputational threats emanating from their political environment, they have an incentive to protect this reputation by signalling their expertise to the external audiences that it deems constitutive to its institutional survival. By actively fostering a reputation for expertise, the technocratic executive can successfully safeguard its legitimacy, and, by extension, enhance its institutional position. The strategic interaction between the technocratic executive and its political environment is schematically displayed in Figure 4.1. In this example, the technocratic executive faces a set of institutional risks (A and B) and a set of societal risks (A  and B). Societal risk B and institutional risks A  and B are perceived by the executive as threats to its reputation. Societal risk A, on the other hand,

Societal risk A

Political environment

Reputational threats

Societal risk B Institutional risk A Institutional risk B

Technocratic executive

Audience A

Signalled expertise

Audience B

Audience C

Audience D

Figure 4.1  Strategic interaction between a technocratic executive and its environment

82  Reinout van der Veer does not present a reputational threat in itself, and therefore does not warrant a response from the executive. In response to these reputational threats, the technocratic executive seeks to signal its expertise to its environment. In this example, its environment consists of audiences A, B, C and D.9 Yet it only perceives audiences B and C as vital to its institutional position, and so it does not purposefully signal audiences A and D. In Figure 4.1, the technocratic executive responds to societal risk A and institutional risks A and B by being strategically responsive to audiences B and C. This, then, is how technocratic executives are responsive to external political threats: not by “pandering” to the preferences of a simple and often volatile majority in a populist sense, but—and similar to the retort given by the climate scientists in response to Trump’s tweet—by “signalling” their competence and expertise to those audiences that are constitutive to their institutional survival. To see how such technocratic responsiveness operates in practice, the following section turns to the case of executive technocracy in the EU.

Three examples of technocratic responsiveness Viewed as a polity, the EU is a most-likely case in which to look for examples of executive technocracy (see also Chapters 1, 2 and 11 of this volume). Since its inception, the EU has traditionally been depicted as a bulwark of technocratic governance that emerged through “integration by stealth” (Majone 2005, Radaelli 1999b). Despite ongoing calls for a reduction of the EU’s democratic deficit, the recent series of financial, economic, debt and political crises that have battered the EU have only bolstered its technocratic credentials (Mény 2014, Mounk 2018, Sánchez-Cuenca 2017, Schmidt 2016). However, it is also a most-likely case in which to observe instances of technocratic responsiveness, as the EU is strongly characterized by the politicization of its legitimacy as a supranational polity. This politicization of European integration consists of three observable components: a rise in the salience of, a polarization of opinion on, and an expansion of actors and audiences engaged in monitoring EU affairs (de Wilde et al. 2016). While surges in politicization are often temporally and spatially bounded (Kriesi 2016), the politicization of the EU has in numerous instances torn through the layers of technocratic insulation and has drawn technocratic actors and institutions into the realm of mass politics. Thus, the cycles of politicization and depoliticization that characterize the EU as a polity create the perfect conditions in which to observe instances of technocratic responsiveness (van der Veer and Haverland 2018b). The following sections discuss three examples of technocratic responsiveness by EU executives in varying levels of reputational peril. The first example consists of a European Regulatory Agency that is challenged by a national regulator with which it shares a core competence, i.e. the conduction of scientific risk assessments (Rimkutė 2018). The  second example illustrates the technocratic responsiveness of the European Commission in its role as the EU’s central economic and fiscal supervisor, which offers invasive recommendations to member

Technocratic responsiveness  83 states in which its legitimacy is heavily politicized (van der Veer and Haverland 2018a). The third example is that of the typically technocratic European Central Bank (ECB), which has nearly single-handedly warded off the sovereign debt crisis through its unconventional monetary interventions. Yet these interventions have also made the ECB one of the most politically controversial actors of the crisis period. The European Food Safety Authority The  emergence of European Regulatory Agencies (ERAs) has profoundly reshaped the European regulatory space since the 1990s (e.g. Levi-Faur 2011, Scholten 2017); in most areas of industry, ERAs now regulate directly or assist the European Commission in doing so. Moreover, they perform most of the enforcement tasks vis-à-vis private actors. Their need to rely on high levels of scientific expertise in order to keep up with industry developments, coupled with their strong political insulation within the EU polity, makes ERAs typical examples of EU executive technocracy. In a recent contribution, Rimkutė (2018) examined how reputational threats affect one of the core competences of ERAs: the conduction of scientific risk assessments. While the goal of these risk assessments is to review existing scientific knowledge on the potential hazards of specific products to people and the environment, her interest was in the regulatory politics of these assessments. Specifically, Rimkutė (2018) considered “how and to what extent regulatory agencies—deemed to be highly technical and primarily scientific bodies— secure their organizational reputation vis-à-vis conflicting audiences by altering their scientific risk assessments”. In search of an answer, she conducted a case study on how two different regulators with the same task handled the scientific risk analysis of the use of the chemical Bisphenol A by the food industry. At the European level, it is the European Food Safety Authority (EFSA) that conducts these assessments. However, some member states have also maintained their national equivalent—in this case, the French Agency for Food, Environmental and Occupational Health and Safety (ANSES). Despite both regulators being asked to perform the same task, they came to different conclusions. The  EFSA  found no direct health risk to consumers, whereas ANSES concluded that stricter regulation of the chemical was needed to better protect the health of some consumer groups. So why did the two regulators arrive at different conclusions? The answer, according to Rimkutė (2018), is that the variation in assessments stems from the different reputational threats each regulator faced. The ANSES was operating in a highly centralized field consisting of strong and mobilized non-governmental organizations (NGOs) and active consumer organizations that possessed recognized authority in the field. This enabled these audiences to issue a powerful reputational challenge to the French agency, making their preferences difficult to ignore. As a result, ANSES opted for a methodology that resulted in a conclusion that reflected high safety standards.

84  Reinout van der Veer The EFSA, on the other hand, operated in a decentralized field where power is diffusely distributed across a heterogeneous set of stakeholders and audiences. Its environment is characterized by conflicting demands and a weak mobilization of interests. As such, EFSA aimed to overcome the various but conflicting reputational threats it faced by asserting its scientific dominance over alternative sources of authority operating in the field. Its conclusion instead reflected high scientific standards, which allowed it to shield itself against charges of inattentiveness to stakeholder interests. Thus, while both regulators relied heavily on scientific expertise, their political use of this expertise differed. This  case study is a clear example of the two logics of responsiveness. Whereas ANSES responded to its reputational threat by ensuring its conclusion was congruent with the preferences of the audiences most vital to its survival, EFSA responded to its diffuse and diverse set of reputational threats by signalling its competence and expertise to the various audiences in its environment. The case study illustrates how technocratic executives use their scientific outputs as political instruments; or, as Rimkutė (2018) concludes, how “the growing emphasis on scientific risk assessments can be explained not only by their analytical function, but also by their relevance as a symbol of rational decision-making”. The European Commission The second example concerns one of the EU’s most important executives: the European Commission. This  multifaceted institution has both legislative and executive competences, but is considered by many to be the “most detached and most technocratic player in the European polity” in either role (Rauh 2016: 3). Many of the EU’s most vital tasks are handled by the Commission, from the initiation of legislation to the external representation of the EU internationally and the enforcement of EU law. However, the power wielded by the Commission also makes it one of the most visible and publicly contested EU institutions, especially during episodes of the politicization of EU affairs—such as during the economic or migrant crises. This  paradox of technocratic insulation and political contestation was the grounds for a study on how the Commission dealt with the politicization of its legitimacy in its role as the central economic and fiscal supervisor of EU member states. Van der Veer and Haverland (2018a) examined how the politicization of European integration in a member state affected the Country-Specific Recommendations (CSRs) drafted by the Commission for that respective member state. These annual recommendations that the Commission offers to member states are based on a heavily expertise-driven supervision cycle and aim to address a broad variety of economic and fiscal “imbalances”. Using a similar theoretical framework to the one presented here, these authors formulated two competing sets of hypotheses that varied by the audiences, and by extension the reputational threats, which would be most important to the Commission. The  first set assumed that as levels of politicization rose, the Commission would be incentivized to pander to the Eurosceptics at the

Technocratic responsiveness  85 domestic level by softening its recommendations. The second set assumed that the Commission would respond to the increasing contestation of its legitimacy by signalling to other member states and market actors that it was a competent and responsible supervisor. In this scenario, the Commission would draft relatively more invasive recommendations for the member state in question as a signal of political resolve. Importantly, a third scenario would mean unresponsiveness on the part of the Commission, as recommendations for more politicized member states would not differ from those in which levels of politicization were lower. Van der Veer and Haverland then employed statistical techniques to examine the content and the quantity of these recommendations, and found compelling evidence for the second scenario. Not only were recommendations to more politicized member states greater in scope compared to those of other, less politicized member states in similar economic positions, they were also less oriented towards social investment than towards social retrenchment. Thus, they conclude that these “recommendations are not solely the product of extensive and objective technical analysis”, and that technocratic, European actors “are not necessarily idle or passive subjects of politicisation, but that they actively develop counter-strategies to respond to developments in their political context” (van der Veer and Haverland 2018a: 18). The responsiveness of the Commission examined in this study is very similar to that of EFSA in the previous example. Both actors signal competence to external audiences, albeit in different ways. This  Commission response to the contestation of its legitimacy is not necessarily the only response the Commission offers, however. In a similar, but far more extensive study, Rauh (2016) examined the Commission’s responsiveness to external signals in its capacity as initiator of legislation in the EU. He found that the Commission actually becomes more responsive to the preferences of a wider audience in that its policy choices become more consumer friendly under higher levels of public awareness. While this certainly hints at the danger of treating an institution as a singular entity, it also shows the importance of executive reputation. Whereas the role of economic and fiscal supervisor requires a competent and responsible Commission, the role of legislator asks for a Commission that is both competent and responsive to citizen demands. The European Central Bank The third example is that of perhaps the most economically significant and politically insulated institution in the EU: the ECB. This technocratic institution has been credited with single-handedly warding off the sovereign debt crisis through its unconventional monetary programmes while showing exceptional political leadership when no other actors would rise to the challenge (Verdun 2017). However, the ECB has also been one of the most politically controversial EU institutions of the last decade. As a high-profile symbol of the EU’s doctrine of neoliberal economic and monetary policy (see Chapter  2 of this volume), the ECB has faced criticism from a wide variety of political and societal actors.

86  Reinout van der Veer This  contestation was most visible during the violent protests in Frankfurt in 2015. Especially since the crisis forced central banks all over the West to embrace much broader mandates, their decisions have become more distributive and have therefore lead to increased political contestation (Fernández-Albertos 2015). There are no systematic academic studies into the responsiveness of the ECB to reputational threats, but there is anecdotal evidence. Firstly, the ECB seems to be aware of the general importance of its reputation as a competent actor for its successful administration of the Euro. As a former economist of the ECB put it: You have to believe that others believe that the central bank knows more than they do. Even if you believe that the central bank does not know more than the market, even that it knows significantly less than the market, then you still follow the central bank if you believe that the others believe that the central bank knows more than the market (qtd in Braun 2015: 383). Due to the idiosyncratic nature of the European Monetary Union (EMU), reputational threats posed to the ECB are more significant than those posed to other central banks. Most notably, the incompleteness of EMU (i.e. the lack of a European fiscal union) means the ECB is placed under the immense pressure of unilaterally holding the Eurozone together (Torres 2013). The general importance of credibility and reputation for central banks, coupled with the amplification of this importance for the ECB specifically, should therefore mean that the ECB is highly sensitive to reputational threats that challenge its credibility. One way the ECB seeks to signal competence is through its external communication. Here, it is first and foremost keen to portray itself as one of the few responsible actors during the crisis period (e.g. European Central Bank 2017). Its communication also became more “competent” as the crisis became more severe: the language of the ECB in its monetary dialogue with the European Parliament, for example, has become more technical in recent years (Collignon and Diessner 2016). This  increase in the technicality of the ECB’s external communication occurred despite the increased interest of “ordinary” citizens in monetary governance during the crisis period, who requested more transparency and more “educational” explanations for what the ECB was doing (Gade et al. 2013: 21). Yet the ECB is also aware of its communicative impact on market behaviour, as communication in positive terms may actually compress bond spreads, whereas negative communication may increase them (Gade et al. 2013). Yet perhaps the most famous example of technocratic responsiveness on part of the ECB was Draghi’s pledge to do “whatever it takes” to save the Euro. His pledge, coupled with his assurance that “it would be enough”, was a major turning point in the crisis and an exceptional display of political leadership (Verdun 2017). However, it was also a clear signal of competence and responsibility as it drove home the message that the ECB was both willing and able to save the common currency. This signal, in turn, both calmed the markets and created the necessary political breathing space for other actors to tackle the crisis. In a similar

Technocratic responsiveness  87 vein, the ECB has hinted it would safeguard financial markets if Brexit lead to severe volatility, and it committed to ending its Quantitative Easing programme shortly after a populist government was elected into office in heavily indebted Italy in 2018. Of course, this evidence remains anecdotal. The  lack of systematic studies on how the ECB seeks to preserve its reputation as a competent and responsible central bank makes it impossible to conclude that the observed mechanisms of technocratic responsiveness seen with EFSA and the Commission also hold for the ECB. Still, the examples offered here seem to suggest that the ECB is sensitive to external reputational threats, and that it responds to such threats by signalling its competence in its external communications—and perhaps also more tangibly in its concrete actions.

Explaining variation: audience attention, preference and power The  previous sections of this chapter offer a generic theory of technocratic responsiveness and review examples of how such responsiveness operates empirically in the contexts of three expertise-driven executives in the EU. While these examples are illustrative, they also remain rather idiosyncratic. The last section of this chapter distils from these examples a set of hypotheses that help further guide scholars in their examination of instances of technocratic responsiveness in other contexts. Following Carpenter (2010) and Rauh (2016), any attempt to explain the responsiveness of technocratic executives should begin by carefully mapping the audiences, and thus the reputational threats, these executives face. Relevant audiences are those that have a stake in, or are somehow affected by, the activities of the executive. Moreover, there needs to be some sort of channel of communication between the audience and the executive through which the former can signal reputational threats to the latter and the latter can signal expertise to the former. For example, in the case of the European Commission such channels of communication are media outlets, but the Commission’s recurrent public opinion survey can also constitute a channel by which publics can signal the Commission. In short, if there is no communication between the executive and a particular audience, the executive will not be incentivized to be responsive to that audience. The first factor that will likely guide the responsiveness of technocratic executives is the level of attention an audience has for the executive’s behaviour, i.e. how salient the executive tasks are to this audience. If the executive’s behaviour is more visible to an audience, this audience will be more likely to issue a reputational threat to the executive, and the technocratic executive is more likely to respond by signalling expertise. In the relatively low-salience case of EFSA, for example, there were a limited number of “usual suspect” audiences for whom EFSA’s decisions were salient. Due to the gravity of the sovereign debt crisis this was different for the ECB. For a few years in a row, the crisis was the most salient issue to a broad set of audiences. While both EFSA and the ECB faced a variety

88  Reinout van der Veer of reputational threats, the levels of attention paid by external audiences were higher for a broader set of actors in the case of the ECB. The second factor that should determine responsiveness is the preferences of relevant audiences. If an audience pays attention but its preferences are congruent with the behaviour of the technocratic executive, the latter faces a limited reputational threat and is in fact responsive to that audience’s preferences in a populist sense. If the behaviour of the executive runs counter to the preferences of an audience, on the other hand, the executive is being unresponsive in a populist sense, and is thus likely facing a reputational threat from that audience. In these cases, it is probable that the executive will be incentivized to signal its expertise. The  examples of EFSA  and the Commission made this very clear. In  both cases, the executives were caught between multiple sets of audiences with divergent preferences for executive behaviour. Thus, the more an audience’s preferences diverge from what the executive is doing, the more likely the executive is to signal expertise. The  third factor is audience power. Even if the behaviour of the technocratic executive is salient to an audience, and even if this audience wishes the executive to behave differently, the audience cannot issue a serious reputational threat without the means to affect the institutional position of the executive. For example, Euroscepticism is only a relevant threat to the Commission’s institutional position because European voters have the power to shape the Commission’s future. Moreover, market actors are only relevant audiences to the Commission and the ECB because collectively they wield the power to alter the economic conditions these executives aim to control. As a third expectation, it is therefore likely that more powerful audiences (i.e. those that are able to issue more serious reputational threats) will generate more incentives for technocratic executives to signal expertise.

Conclusion To summarize, this chapter argues that technocratic executives in liberal democratic systems may not be as unresponsive as they may initially appear. Whereas expertise-driven executives may not always pander to the people in a populist sense, they can be responsive to reputational threats issued by key audiences in their political context. Technocratic executives may respond to such threats by signalling their expertise in order to bolster their reputation as competent and responsible actors. Subsequently, the chapter offers three illustrative examples from technocratic executives in the EU context, which portray how such technocratic responsiveness works in practice. The previous section condenses three sets of hypotheses from these examples that will help guide future research on the subject. Yet the unravelling of the mechanisms through which technocrats balance the need for responsiveness and responsibility leaves open the more important question of how this reflects on the overarching question posed in this volume: does technocratic responsiveness mitigate or aggravate the technocratic challenge to democracy? With reference to the responsiveness–responsibility

Technocratic responsiveness  89 dichotomy discussed at the beginning of the chapter, technocratic responsiveness implies purposefully and visibly signalling responsibility in response to political (in the form of institutional and reputational) threats. Even if such signalling is merely symbolic, technocratic executives need to be willing to back such signals with tangible actions if they wish these signals to be credible. Viewed this way, such technocratic responsiveness may actually have a positive consequence in that a focus on responsibility caters to the preferences of future legislatures, but also to broader sets of actors who are not represented by an actual or constructed simple majority. Thus, the strengthening of an executive’s reputation as a competent actor also aids the responsibility credentials of democratic political systems. However, the political strategy of technocratic responsiveness also creates new sets of risks for technocratic executives. Most notably, it may bring about a different set of political threats to the institutional position of the executive, and may in fact amplify the technocratic challenge to democracy. Evidently, signalling expertise may run counter to the preferences of a majority of citizens in society. In the example of the Commission, signalling expertise by being stricter with Eurosceptic member states may further strengthen anti-expert or anti-EU sentiment among publics, which may ultimately increase the very political threat the strategy is responding to. This may create an undesirable feedback loop, where stronger reputational challenges are met with even stronger acts of technocratic responsiveness. Strategies of technocratic responsiveness may also open insulated technocratic executives up to the critique that their actions favour an elitist, undemocratic agenda. There  is a real danger that any behaviour that is linked to the responsibility dimension is denounced by political outsiders as favouring the political and economic status quo. At the same time, the unjustified concealment of any status quo-favouring behaviour as “responsible action” by political elites is equally problematic as it harms the responsiveness of government to citizen demands. Thus, these strategies may ultimately strengthen the populist challenge to democracy by weakening the credibility of political appeals for responsible (and therefore at times unresponsive) government. Returning to the issue of climate change, President Trump often counters the retorts offered by climate scientists with a variation on the following statement: “Look, scientists also have a political agenda” (Friedman 2018). This points to a more fundamental problem in that the political use of science and expertise degrades trust in science and expertise itself. This trust has already decreased in recent decades (Motta 2018). Technocratic responsiveness as a political strategy will further strengthen this development, as it explicitly places at odds the populist and technocratic sources of legitimation: will and reason (Caramani 2017). Trump’s quote suggests that reason has its own will, and that this will is different from the will of “the people”. Yet will and reason are not irreconcilable opposites (see Chapter 1 of this volume). In liberal democracies especially, they should (and often do) go hand in hand synergistically to create imperfect yet optimal political solutions to societal problems.

90  Reinout van der Veer

Notes 1 While this definition does not necessarily exclude technocracy as a decision-making method, the prerequisite of discretion means this chapter more strongly speaks to technocracy as a regime and/or government (see Chapter 3 of this volume). This is because as long as an executive enjoys discretion in shaping political outcomes, and different outcomes yield different “winners” and “losers” (see Chapter 1 of this volume), the executive has some degree of control over both means and ends. 2 For a more detailed discussion on technocracy and temporality, see Chapter 1. 3 The political construction of such ad hoc simple majorities, whether truly existing or rhetorically constructed by the populist, is also possible in the most proportional political systems. 4 Technocratic ideology does not preclude political conflict, but instead assumes that political conflict is overcome once the optimal technical solution is found. 5 This does not refer to central banks that are (in)directly accountable to government or the legislature, such as the Bank of England under the Westminster system. 6 See Chapter 2 for an economic argument for central bank independence. 7 With reference to technocratic executives, “regulation” is used in this chapter as a verb: whereas some technocratic executives may produce regulation, many if not all are regulating (i.e. controlling) some part of our societies. 8 As opposed to its uses in Chapter 3, politicization here refers to the empirical phenomenon of an increase in the salience and political contestation of the executive’s actions amongst observing audiences. 9 In  practice, the difference between audiences and (especially institutional) risks is less clear cut, as audiences can produce both risks and reputational threats themselves. The  example of the European Commission below provides an insightful example of the inherent endogeneity between audiences and (institutional) risk.

5 Measuring technocracy Eri Bertsou and Daniele Caramani

Introduction The technocratic ideal has a long tradition ranging from the classical world to the Industrial Revolution and the Taylorist movement of the 1920s and 1930s. Although it has seldom been implemented in its pure form of “experts exercising political power”, the scientific ideals it champions have become part of many democracies. In the second half of the twentieth century, the increasing complexity of policymaking, the expansion of supranational governance and the growing interconnectedness of economic systems created increasing demands for expertise in governance and the depoliticization of policymaking. More recently, the first two decades of the twenty-first century have seen the appointment of technocratic governments in Italy, Greece, Hungary, the Czech Republic and Romania in response to economic or political crises.1 At the same time, the criticisms mounted at supranational and intergovernmental institutions, such as the European Union (EU) or United Nations (UN) bodies, show no signs of dissipating following the financial crisis and have led to a backlash of populist and anti-intellectual politics. These developments have propelled questions regarding the role of expertise and independence in politics into the public spotlight. Such questions are timely and will linger in the years to come, as polities around the world try to balance their wish for more responsiveness to the people with the need for expertise. These discussions have instigated scholarly contributions surrounding the theoretical standing of technocracy and its relationship to representative democracy, which are also addressed in the previous chapters of this volume.2 So far, however, far less emphasis has been placed on a systematic empirical approach to the operationalization and measurement of technocratic elements within representative democracies. This  is a paradox, since some of the most important questions regarding the role and effect of technocratic politics within democratic systems are empirical. Unlike other challenges to representative democracy— such as populism, political disaffection or the decline of party membership— empirical research on technocracy cannot yet rely on a solid framework that allows the cross-sectional comparison of levels of technocratic influence in different political systems, the longitudinal trend of technocratic elements over time

92  Eri Bertsou and Daniele Caramani and, ultimately, the testing of hypotheses regarding the increase and effects of “technocratization” in established democracies. Instead, most of the empirical literature on technocracy has relied on single case studies and qualitative indepth accounts of cabinets and countries (Centeno and Silva 1998, Hanley 2013, Neto and Lobo 2009, Pasquino and Valbruzzi 2012, Protsyk 2005a, 2005b). We believe that the fundamental questions regarding technocracy and democracy cannot be answered without empirical analysis. Empirical analysis should include questions concerning the level of technocracy entailed in our democracies, its effects on policy and governance, how much technocracy do people, elites or given parties want and how much technocracy do they think is desirable for democracies. At all the different levels identified in the Introduction to this volume, such questions require empirical investigation. This chapter offers a roadmap for the systematic categorization, operationalization and possible empirical measurement of technocratic elements in democratic systems. Using the analytical framework developed in the earlier chapters of this volume, it gathers, reviews and adds to efforts of empirical analysis of technocratic politics.3 It presents indicators and measures at different levels of the political system: individual (citizens and representatives), organizational (political parties and other political organizations), institutional (state institutions) and systemic (the relationship between politics and governance agencies). At  each level of analysis we consider a variety of methods, including behavioural, discourse and aggregate methods. The ultimate goal is to provide a basis for comparative, cross-sectional and longitudinal analysis of technocratic elements in different political systems.

The relevance of technocracy in empirical research The question of technocracy and its contested relation to democracy entered the public debate at the height of the financial crisis in Europe, when democratically elected governments in Italy and Greece were replaced by the technocratic cabinets of Monti and Papademos (Culpepper 2014, Pasquino and Valbruzzi 2012, Pastorella 2016). The  common narrative when discussing these technocratic appointments, as with other technocratic cabinets, referred to the handling of some type of crisis when “normal” processes of representative democracy can no longer offer a swift and effective solution to the problems the country is facing (Wratil and Pastorella 2018). However, the process of depoliticization of governance had been well underway for years across established democracies, whether by choice or necessitated by globalization, supranational coordination and an interconnected financial system (as pointed out in the Introduction to this volume as well as in other ­chapters; see also Fernández-Albertos 2015 and Fischer 2009 for overviews). Early scholars of the EU stressed the bureaucratic and, indeed, technocratic nature of this emerging political entity and pointed to the inherent democratic deficit of decision-making processes that are immune to electoral competition, party politics and democratic legitimation (Majone 1994, Moravcsik 1998, 2002, Radaelli 1999b,

Measuring technocracy  93 Sánchez-Cuenca 2017). Other policy paradigms, from larger-scale economic interventions in Latin American countries to policy programmes in the United States, followed in the same footsteps of depoliticization, expert governance and its independence from political control (Centeno 1993, Dargent 2015, Fischer 1990). Beyond these most visible instances of technocratic politics, democratic political systems more generally have always needed to harness expertise and independence to function effectively. In many cases, the depoliticization of decision making is democratic, and even desirable, even though it circumvents the link between the citizens and the representatives they elect to enact their preferences (Pettit 2004). Technocratic elements are profuse in democratic systems, but without the right empirical measures it is impossible to know “how much” political systems are technocratic. It is also difficult to identify whether technocratic influence is increasing over time and to assess its effects on political outputs. All this is necessary before we can address more complex questions regarding the functioning of democratic systems, such as whether the “technocratization” of politics causes a populist backlash, with demands for more responsiveness and the repoliticization of governance, and what are the conditions under which technocratic elements are a “corrective” that rebalance the responsibility role of representative government. There may be parts of the political system that are better served by independent, expert-based and “non-majoritarian” entities. Our focus in this chapter is on key players of representative democratic systems, which are traditionally thought of as the locus of decision making in representative democracy: parties, executives, parliaments, politicians and the citizens that elect them.

Theoretical basis for the study of technocracy The concept of technocracy has been the object of scholarly attention in the works of Meynaud (1969), Fischer (1990) and Centeno (1993), with renewed interest from Rosanvallon (2011), Bickerton and Invernizzi Accetti (2017) and Caramani (2017). In  its ideal form, the term technocracy signifies the rule or exercise of political power by a group of experts. In this volume, we follow a more inclusive and flexible definition of technocracy as a form of power, legitimacy, representation and decision making that stresses the role of expertise, skill and unattached interest. The  aim of technocratic politics is the identification and implementation of optimal solutions to societal problems, which will ensure the long-term prosperity of the political community. It champions responsible governance and rests on a belief that objective solutions exist and can be reached through impartial analysis, scientific reasoning and expert knowledge. Therefore, the source of power and legitimacy is not  popular selection, but knowledge and expertise. Implicit in the concept of technocracy is the belief that governing a political community is a complex and taxing task that requires dedicated experts with skill and experience to make the right decisions. Other actors, such as citizens or elected officials, may be unable (lack the expertise) or unwilling (serve other interests) to govern responsibly.

94  Eri Bertsou and Daniele Caramani Technocracy is opposed to the ideal of party-based representative democracy, where decisions are reached through a confrontation between competing interests and preferences.4 In this sense, technocratic politics is anti-pluralist and monolithic, as it recognizes only one interest—the long-term prosperity of the entire community—and rejects all partisan, class or other group-based conflict as an impediment to this goal. Nevertheless, there is a claim to representation. What is being represented is precisely this “long-term prosperity of the entire community”. In terms of representation, technocracy sits at the extreme end of trustee in the “trustee−delegate” continuum. Moving from the definition to the operationalization of technocratic politics, we identify the following three dimensions: •

•

•

Elitism: Technocracy identifies an elite group, based on knowledge, expertise, academic credentials, intellect and know-how, and separates it from ordinary citizens. This elite is considered to “know best” and should guide society responsibly, with the long-term goal of prosperity. This elite sits in sharp contrast to the existing political class, political parties and any nonstrictly meritocratic processes and institutions. Anti-pluralism: Technocracy adheres to a monolithic view of society. It criticizes ideological commitments and partisan interests that seek to benefit sections of the community rather than the community as a whole. Technocracy does not recognize political conflict between societal groups on the basis of sectoral, class or minority status and considers parties and interest groups as damaging to the prospects of societal prosperity.5 Technocracy is antipluralistic in a political sense, that is, anti-partisan and sceptical towards elections, which are perceived as popularity contests. Whereas populism separates the world in “good” and “evil”, technocracy opts for a scientifically informed dichotomy between “right/correct” and “wrong/mistaken”. Scientific approach: Technocracy is based on the belief that there are objective, neutral, non-ideologically committed experts, who are able to enact the best solutions to governance problems after evaluating relevant evidence and facts. Technocratic politics follows a positivist paradigm. There is a belief in the existence of an optimal solution or truth, which can be discovered through careful and objective analysis of scientific evidence. It  prioritizes output, efficiency and optimal outcomes over other types of legitimacy and views society as a machine with many moving parts that need to operate effectively. This scientific approach complements the aforementioned elitist, anti-political and anti-pluralist dimensions by emphasizing the role of expert knowledge, political neutrality and problem-solving capacities.6

From the definition of technocracy given above and the dimensions that derive from it, technocratic elements become apparent at different levels of democratic political systems and across a variety of actors and processes. In the following section, we identify these elements and provide a map of possible methodological tools to empirically measure technocracy in the most important areas of political research.

Measuring technocracy  95

Towards a typology of empirical measures of technocracy Given its complex, multidimensional and normative character, it is perhaps not surprising that the operationalization of the concept of technocracy from a comparative and longitudinal perspective has been elusive. Studies have mainly focused on a single aspect at a time, such as in-depth case studies of specific countries and cabinets (Hanley 2009, Pasquino and Valbruzzi 2012, Stegmaier and Vlachová 2011).7 Without a doubt, case studies still offer a wealth of information on the specific characteristics, processes and effects of technocratic politics. Yet the goal of this chapter is to propose a typology and offer suggestions for the operationalization of technocracy that can lead to comparative analysis. It reviews existing research efforts and proposes new avenues to “quantify” technocracy across different layers of the political system, and by employing diverse methodologies. We proceed in two steps. First, we provide a typology for research into technocracy for a given level/method combination. Second, we discuss indicators and measures for each, with examples wherever possible. To structure the map of indicators for the measurement of technocracy in an encompassing and systematic way, Table 5.1 distinguishes three main levels of analysis and four methods of analysis to investigate technocratic politics in democratic systems. We label Table 5.1  A typology of research in technocratic politics Method of analysis

Levels of analysis Individual (elites, non-elites)

Aggregate (organizationalinstitutional)

Systemic (nationalsupranational)

Sociological/composition analysis (Data: individual characteristics)

Who is a technocrat? (1) expertise (2) independence

How many technocrats are in a given group?

n/a

Discourse/text analysis (Data: texts)

Who has a Which group has technocratic a technocratic discourse? discourse? How technocratic How technocratic is their discourse? is their discourse?

n/a

Behavioural/ attitudinal analysis (Data: survey questionnaires, experiments)

Who prefers/ supports/trusts technocrats?

Which group prefers/supports/ trusts technocrats?

n/a

Procedural analysis (Data: constitutions, treaties, negotiations, signalling)

n/a

How prominent/ powerful are technocrats in a given group?

How prominent/ powerful are technocratic institutions in a given system?

96  Eri Bertsou and Daniele Caramani levels as “individual”, “aggregate” and “systemic”, and focus on four types or methods of analysis: sociological, discursive, behavioural and procedural. Table 5.1 is populated with the types of questions addressed by each method/ level combination. When one moves from the individual level to the aggregate level this would usually involve simply an aggregation of the analysis carried out. For example, identifying who is a technocrat is an individual-level analysis, based on the characteristics of a given figure, which is necessary before identifying technocratic cabinets or legislatures at the aggregate level.8 Similarly, one can analyse the discourse of a given actor at the individual level, but also group together discourses of actors who represent the same group—a given political party, for example—to conduct research at the aggregate level. Thus, one may find a different composition by parties or cabinets, or different discourses and different attitudes for expert-driven governance among members of different parties. Sociological, discourse and behavioural methods of analysis can be applied at an individual and an aggregate level, but the upward aggregation and breakdown of data by organization and institution also stops at the last level of analysis we identify: namely, the systemic level. The relationship between institutions—in particular, the degree of independence of regulators and agencies from political institutions acting on a democratic mandate—cannot be derived from behavioural or discourse indicators. The same applies to the degree to which national institutions and representatives are constrained and depoliticized by their being embedded in supranational agreements. Again, this cannot be derived from aggregate composition data from lower levels of analysis. Where the interactions between agents are most informative—that is, where procedural analysis can take place—one necessarily needs to move away from the individual level and look at groups and institutions. Individual level Sociological The first obvious task in the analysis of technocracy involves identification of “the technocrat” (see, for example, Chapter 7). This applies to analyses of elites, be it political candidates, leaders, cabinet ministers, Members of Parliament (MPs) or members of the bureaucracy. Throughout this volume we understand “technocrats” as independent experts active in politics. Expertise can be observed and measured on the basis of individual characteristics, background and track record. Educational attainment, previous professional experience and political involvement, and other relevant credentials can be used as evidence of expertise. Independence or outside status is more controversial. While political involvement is particularly relevant alongside attestations of expertise in establishing to what degree an individual is independent from political affiliations, here there are diverse views about the “outsider status” of elites. Ideally, political independence would mean no prior political experience or political ties.9 A long professional

Measuring technocracy  97 career in a relevant sector for governance is by default inversely related to one’s political ties. However, in reality, the lines are more blurred. It  is possible for independent experts to align themselves with a political party, start their own political movement, or become so prominent in the politics of a nation that they do become experienced politicians. Camerlo and Pérez-Liñán (2015) noted the importance of separating technocrats from simple “outsiders” or people appointed from “personal networks”. Outsider status can be used by political entrepreneurs, especially when grievances are mounting among citizens, but without a proven track record of professional expertise in a relevant sector of governance (corresponding to the “elite” and “scientific approach” dimensions of technocracy), such individuals should not  be classified as technocrats. Table  5.2 (taken from Camerlo and Pérez-Liñán 2015) shows the categorization of elites according to political affiliation and expertise. In  their analysis of technocratic governments, McDonnell and Valbruzzi (2014, see also Chapter 6 in this volume) formulate the three criteria for assigning the technocratic label to the leader of the executive: (1) never having held public office under the banner of a political party, (2) not being a member of any political party and (3) possessing recognized experience and expertise (2014: 657–58). This  sets a high threshold for inclusion into the “technocratic club” and does not account for the presence of relatively more or less technocratic elements to be found among politicians. We argue that expertise is more important than political affiliation, especially given the fact that independent experts often decide to affiliate themselves with a political party, or create their own, at a later stage in their careers.10 An alternative approach was followed by Despina Alexiadou, who in her classification of cabinet ministers includes the category of “ideologues” for individuals with firm stances on policy direction that trumps office-seeking and vote-seeking considerations (2015, and Chapter 7 in this volume). This includes trade union officials, who have experience in wage-bargaining, workers’ rights and union–employer relations, and a strong conviction regarding economic and social policies. Similarly, many technocratic appointments (by elected or appointed heads of state) are based on the relevance of experience and expertise. Macron appointed a former Olympic champion as Minister of Sport and a doctor as Minister of Health. Similarly, Monti appointed an academic and university Table 5.2  Classification of elite profiles Party affiliation

Yes No

Expertise With expertise

Without expertise

Partisans Non-partisan technocrats

Partisans Non-partisan outsiders

Source: Camerlo, M. and Pérez-Liñán, A., Comp. Polit., 47, 318, 2015.

98  Eri Bertsou and Daniele Caramani dean as Minister for Education, and a naval officer and NATO expert as Minister of Defence. Though Alexiadou’s classification is not aimed at identifying technocratic elements, there is a large overlap and all technocrat ministers fall within the category of “ideologues”. Further, her definitions of expertise and experience draw attention to an existing bias in the types of knowledge usually considered to be technocratic. This bias is two-fold. First, it is an ideological bias in favour of economically liberal views and training. Second, it is a content bias in favour of the natural sciences, engineering and economics (as it is a-ideological, as the Introduction discusses). There is no inherent connection between technocracy and liberal economic or capitalist views; nevertheless, in practice, institutions that sought to depoliticize policy in order to reduce conflict have been primarily supranational, financial and regulatory in nature and have followed liberal economic principles (Fourcade‐Gourinchas and Babb 2002). Economic and financial expertise has become synonymous with experience in international financial institutions (IFIs) and their requirement for “adjustment”. Yet an accurate assessment of expertise should reject this bias. The second bias encountered frequently in the public discourse on technocracy is in favour of the formal and natural sciences as appropriate backgrounds for and evidence of expertise. In  this case, there is a fundamental connection between technocracy and the hard sciences. It emphasizes the technical nature of governance, and its positivist approach to the social world—that is, the belief that there is one true, right solution to a problem which can be discovered through unbiased scientific analysis. This is a much closer fit to the paradigm of mathematics, engineering, physics and other natural sciences. However, for the purposes of assessing expertise at the individual political level, the “discipline” or “content” of one’s experience and knowledge needs only to pertain to the governing task assigned. Discursive An alternative method of analysis at the individual level would be to identify technocrats based on the discourse used. Given that technocracy entails a legitimation strategy of political decisions, the communication of political elites offers a promising field for the study of technocracy. Technocracy represents a value-system that can be found in public discourse. This valuesystem is in line with the dimensions of elitism, anti-pluralism and scientific approach presented earlier, but also additional elements to be found in oral and written text. The  type of texts that can be analysed at the individual level are speeches by political leaders and other text they produce, such social media broadcasts (tweets) and content on personal websites. Texts can be analysed in either a qualitative or a quantitative way (Poblete 2015, Rooduijn and Pauwels 2011), using human coders or programmes for automated analysis. Empirical possibilities include holistic coding (Hawkins 2009) or more precise hand-coding of

Measuring technocracy  99 each paragraph, sentence or quasi-sentence of texts. Other possibilities include computer-based coding, which produces numerical results of the occurrence and frequency of sets of words, or machine-learning methodological approaches, whereby the computer is instructed how to code based on a training set of text and can then be given a new corpus to analyse (Grimmer and Stewart 2013). To our knowledge, there has been no such study of technocratic politics using text and discourse prior to this volume. In Chapter 9, Nava, Ángel Centeno and Liu study the Brexit campaign using a top-down methodology of structured topic modelling. In this chapter we present an alternative approach that would be more in line with human coding according to the presence/absence of relevant characteristics of technocracy. Table 5.3

Table 5.3  Codebook for the analysis of technocratic discourse (excerpt) Dimension 1: Elitism

Dimension 2: Antipluralism

Dimension 3: Scientific approach

Dimension 4: Outputoriented

Dimension 5: Technical style

Positive: Praise knowledge elite, critical of ordinary citizens and politicians. Implicit or explicit higher value attached to this group, as people “who know best” and as those who can and should guide the society. Negative: Praise of ordinary citizens, politicians or representatives of the people, critical of unaccountable elites, experts. Positive: Reference to the welfare of society as a whole, criticism of ideological and partisan interests, special interest groups and societal struggle between different groups. Negative: Reference to interests of specific groups, sectoral (ex. workers in heavy industry), ethnic groups, social class, and reference to struggle between social groups, criticism of the “illusion” of what is the best solution for society. Positive: Reference to objective, neutral, non-compromised decision making, evidence-based policymaking and critical of voices that question facts or the scientific method. Negative: Denigration or doubt of factual evidence, reference to subjective or compromised decision- and policymakers, doubt of the complexity of social problems. Positive: Praise of efficiency, output, growth for the entire society and reference to what works, what provides optimal outcomes. Disregard for criticism that refer to the importance of procedural questions and non-output-related values. Negative: Reference to procedures and non-output-related “achievements”, criticism of principles of efficiency, output, and optimization. Positive: Use of technical jargon, facts, figures, sources (many legitimate sources), dry language, speech that does not stir an emotional response. Clean, precise and sophisticated language. Negative: Use of layman’s terms, inconsistent use of sources or misrepresentation of figures, emotive and value-laden speech that promotes and gives agency to the speaker (active voice, reference to “I” or “We”).

Source: Codebook developed by Bertsou and Caramani.

100  Eri Bertsou and Daniele Caramani provides an excerpt from the codebook that we developed during a graduate research seminar at the University of Zurich (the full codebook is available upon request). The codebook is based on the three dimensions of technocracy presented earlier (elitism, anti-pluralism and scientific approach), and is supplemented by two more elements that are specific to discourse: output focus and technical style. Texts that score highly on these five dimensions represent ideal examples of technocratic discourse. While this codebook was developed and tested using human coders, it can provide the basis for more detailed instruments for text analysis.11 A similar analysis has been carried out by the project NCCR Democracy at the University of Zurich, in collaboration with communication scientists, in the analysis of the appeal of populist ideas. The database includes a number of items that can be used for technocracy. Numerous discourse strategies can be derived by being the opposite of populism, such as shifting the blame to supranational institutions and financial elites, or expressing closeness to the people. Conversely, some strategies are similar to populism, such as a monolithic view of the people or a negative contextualization of mainstream politics. Also, the style can be the opposite of populists, such as the use of data and statistics, or not relying on common sense and black-and-white rhetoric, dramatization and emotions, but rather on academic jargon. Clothing has also been coded in this project. In terms of justification of statements, economic justifications or previous agreements are technocratic features, while people’s will and common sense are not. Argumentation is based on competence and causality. The targets of the statements do not identify elites as negative.12 Behavioural A  third methodological approach to study technocracy at the individual level relates to the preferences, attitudes and behaviour of individual actors. This represents a particularly fruitful avenue for research, and it can apply to the “supply side” of politics (political elites and bureaucracies) as well as the “demand side” (citizens). In the former case, we can analyse MPs’ preferences or attitudes and how they interpret their representative “role”, for which they are accountable to citizens. In the latter, we can study the demand for independent expertise coming from citizens. We therefore discuss the two separately. SUPPLY SIDE

Putnam (1977) attempted to operationalize and measure “technocratic mentality” among civil servants in one of the earliest quantitative empirical analyses at the individual level. Apart from categorizing civil servants in terms of their training, he compiled survey evidence regarding their approach to their job tasks to estimate levels of political neutrality and tolerance for politics. Already in 1977, he found that although the threat of technical rule had not materialized, the rise of the hybrid politician–technician was a major trend in administrations and bureaucracies across capitalist and communist industrialized countries.

Measuring technocracy  101 While this line of research was not followed up, it can still provide fruitful data for the study of technocracy among elites. Apart from a “technocratic mentality” index, the trustee model of representation provides a solid theoretical basis for analysing responsible governance, a fundamental aspect of technocratic politics. Officials may have been appointed to their posts or selected through the process of elections by the represented, but a trustee role implies independence and the absence of constraints during their remit. Under the trustee model of representation officials are entrusted with the task of governing, because of their knowledge, expertise and good character, and their responsibility is to carry out that task to the best of their ability. Much like technocrats strictly speaking, part of their legitimacy originates in their superior knowledge and skill justifying a nonresponsive action in favour of a responsible one.13 The closer an official follows the trustee model of representation—that is, making decisions not on the basis of a mandate, partisan or other, but on what is best for the community following her or his best judgement, knowledge and experience—the more technocratic their role. It is important to point out, of course, that although the trustee model of representation rests on the premise that citizens do not have the time and/or capacity to identify what is best for their community (and hence this task should be entrusted to individuals with the requisite skill and dedication), there is no guarantee of expert representatives. DEMAND SIDE

Finally, investigations of technocracy have surfaced recently in an effort to understand why citizens accept—or even prefer—the appointment of external, independent decision makers at the helm of their country.14 A prominent example is that of the concept of “stealth democracy” (Hibbing and Theiss-Morse 2002), which measures citizens’ preferences for less popular involvement and more effective decision making carried out by unelected experts or business people. The authors found that a large chunk of Americans welcomed a more detached and efficient way of governing that can bypass disagreement and division of ordinary politics. The  first empirical study of citizens’ technocratic attitudes in a comparative perspective (Bertsou and Pastorella 2017) relied on a survey measure of technocratic decision-making preferences as a proxy for the underlying trait of technocratic attitudes. Their analysis showed that citizens’ trust in representative political institutions (political parties, national parliament and government) and their beliefs about the benefits of democratic governance have a negative effect on technocratic preferences overall, pointing to an understanding of the technocratic elements of EU governance.15 This  analysis also revealed large cross-national variations in the attitudes of citizens towards technocracy, with Central and Eastern European countries being much more favourable towards expert decision making, and some Southern European and Nordic countries highly critical of technocratic politics. These differences appear to persist over time. Figure 5.1 shows preferences for technocratic decision making across established democracies included in two

102  Eri Bertsou and Daniele Caramani 100% 90% 80% 70% 60% 50% 40% 30% 20% 0%

2008 2010-2014 2008 2010-2014 2008 2010-2014 2017 2008 2017 2008 2017 2008 2017 2008 2010-2014 2017 2008 2010-2014 2017 2008 2010-2014 2017 2008 2010-2014 2008 2010-2014 2008 2010-2014 2017 2010-2014

10%

Great CyprusEstonia Germany Britain Greece Italy Netherlands Poland

very bad

fairly bad

fairly good

RomaniaSloveniaSpain Sweden USA very good

Figure 5.1  Attitudes towards experts in governance. (Notes: Data taken from EVS 2008, WVS 2014 and own survey. The question reads: “Do you think having experts, not government, make decisions according to what they think is best for the country is a very bad/fairly bad/fairly good/very good way of governing this country?”)

waves of the World Values Survey (WVS) and European Values Survey (EVS) between 2008 and 2014 and the authors’ own survey from 2017. Measuring the underlying traits that manifest as technocratic preferences requires a more nuanced approach. In an ongoing research project, we have devised a scale to measure the latent construct of technocratic attitudes, through multiple survey items that tap into different dimensions following the theoretical understanding of technocracy presented above (Bertsou and Caramani 2017, forthcoming). Having a detailed, theoretically driven and empirically valid measure for technocratic attitudes allows comparison with other attitudes, attributes or individual and group-level characteristics, such as ideology, education or voting. The scale is validated across different national contexts and can therefore provide a measurement tool for scholars of political behaviour, political psychology and other fields of political science.16 Technocratic preferences can also be measured at the individual level without having to delve deeper into attitudinal concepts. The presence and influence of independent experts are a reality in many democratic systems, with agencies, institutions and political players considered to be “more technocratic” than others. It  is therefore possible to gauge citizens’ support and trust for those players. Costa Lobo and McManus investigate support for technocratic institutions in managing the financial crisis in Europe in Chapter  11 of this volume. Even without concrete real-life political dilemmas, conjoint experiments offer another potential avenue to study the conditions under which and areas in which citizens prefer to delegate power to technocrats. This  points

Measuring technocracy  103 to a promising area of research that investigates the intersection of supply and demand of technocratic elements, and citizen responses to increased or decreased technocratic governance.17 Aggregate level We use the term “aggregate” to refer to the types of analyses that look at both organizational and institutional political groups. This  would include political parties, interest groups, cabinets, legislature, agencies and the civil service. Sociological, discourse and behavioural analyses can all be used at the aggregate level, facilitating comparisons between groups and over time. Once individuals at the elite level have been classified according to a technocracy scale, it is possible to compare the composition of, for example, different parties, ministries or whole cabinets between countries and/or over time. Parties can be analysed according to their reliance on “technocrats”, for example, as opposed to grassroots volunteers, or the technocratic discourse of the legislature can be traced over time to show whether it increases during times of financial crisis. This is where we find most existing attempts to measure the technocracy of cabinets.18 McDonnell and Valbruzzi (2014) provide an operational definition of a technocrat (necessarily at the helm of a technocratic government) and, further, four groups of such governments based on their remit and composition. Whereas their definition of “a technocrat” is dichotomous, they separate non-traditional technical governments based on their degree of technocracy. This  typology includes non-partisan caretaker governments, technocrat-led partisan governments and full technocratic governments. They  identify 24 governments in Europe since the end of World War II, but only 6 that are fully technocratic in their remit and composition, leading them to conclude that “European governments remain, overwhelming, duly-mandated party governments” (2014: 666).19 An analysis focused solely on the classification of governments led by a technocrat leaves a wide range of technocratic elements at the level of cabinets unexplored. Whereas full technocratic governments may be rare, independent ministerial appointments and the reliance on technocrats occurs more often in the absence of such cabinets (Neto and Strøm 2006, Pasquino and Valbruzzi 2012). The concept of “government partyness” is closely and inversely related to the degree of technocracy in a given cabinet and can be extended to other organizational levels (for example, the extent of partisan appointments in the bureaucracy). The partyness of government was conceptualized by Katz (1986) as the importance of the political party in its governance role. Though its operationalization varies, one aspect of partyness is the percentage of non-partisan ministers appointed in each cabinet (Katz and Mair 1995, Kopecký et al. 2012). Figure 5.2 follows this calculation by Pasquino and Valbruzzi (2012) and shows the partyness of government among Western European countries from 1945 to 2012. At  the aggregate level, discourse analysis offers a fruitful avenue for the study of texts produced by political organizations, party manifestos being the best example. Other texts produced by political parties and other organizations,

104  Eri Bertsou and Daniele Caramani

0,96 0,98 0,99

0,98 0,83

0,96 0,86

0,92

0,99

1

0,93

0,98 0,97

0,97

1

0,98 0,99

1

0,75 0,62

Au st Be ria lg i De um Fr nm an ar ce k F IV Fr (1 inla an ce 946 nd V –1 (1 95 957 ) 8– 20 G e 12) rm an Gr y ee c Ice e la nd Ire la Ita nd ly I( Ita 19 Ita 4 ly ly II ( 5–1 19 99 94 3) – Lu 201 xe 2) m Ne bu r th er g la nd No s rw Po ay rtu ga l Sp ai Un n S ite we d d Ki en ng do m

1 0,9 0,8 0,7 0,6 0,5 0,4 0,3 0,2 0,1 0

Index of partyness of government

Figure 5.2  Partyness of government in Western Europe, 1945–2012. (From Pasquino, G. and Valbruzzi, M., J. Mod. Ital. Stud., 17, 616, 2012.)

interest groups (trade unions or NGOs) or even social movements, which act more or less according to a logic of advocacy representation, can be analysed. Party manifestos, parliamentary debates and official documents would allow for within- or cross-country comparative and longitudinal analysis according to the technocratic dimensions mentioned earlier. In this way, political scientists can study the degree of change over time (technocratization) and compare the different legislative branches. Systemic level The last level of analysis is the systemic level, where we focus on the relationship between non-elected or apolitical agencies on the one hand, and elected, representative and majoritarian institutions on the other. The question of interest at this level is: how powerful and independent are non-elected governance bodies that are not embedded in an electoral representation circuit? In other words, we need to examine the degree to which decision-making processes are “depoliticized” and non-majoritarian—i.e. not submitted to a public confrontation of alternatives and, ultimately, a direct or indirect electoral choice. The thesis of technocraticization in recent years has gained traction precisely as the number and independence of governance agencies, authorities, regulatory bodies, think tanks, rating agencies, etc., has increased. It  is highly challenging to empirically document such an increase, and it is here that one finds most qualitative (case study) types of analyses on the procedures of decision making, processes of displacement of competences beyond the politically representative checks. This dislocation of “authority” has been documented at the national level in matters of policy, oversight and public management (Fischer 1990), but has

Measuring technocracy  105 been more evident in the pooling of previously national political competences to depoliticized supranational bodies. As discussed, the EU has contributed to a great extent to this depoliticization at the systemic level, as political power has been transported to the “Eurocrats” of the European Commission or the European Central Bank (ECB). The key aspects of this technocratization can be measured in terms of “limitations” set on politics, whereby the “choice” element between alternative programmes presented to the electorate is taken away.20 Such limitations on choice appear not only when decision making is “outsourced” to national independent bodies or supranational organizations, but also in respect of international relations, with the decisions by some countries limiting or even discarding room for manoeuvre in other countries—in particular, when linked to financial loans. This is where globalization—in the sense of a growing interdependence between countries (especially when institutionalized by bodies such as the International Monetary Fund  [IMF] and the World Bank)—leads to a “hollowing out of democracy”. A  similar example concerns the “conditionality” placed on some countries by groups of other countries to access membership or access loans.21 Where political choice is limited, democracy is diminished and decision making becomes technocratic. However, what this vision does not  consider—and which is relevant for the empirical measurement of technocracy at the systemic level—is whether or not accountability in these independent bodies can be conveyed through different mechanisms. To what extent are these agencies, bodies and authorities accountable to politically elected institutions? How strong and “last instance” is the principal’s control over these bodies?22 The empirical analysis required to answer such questions is obviously largely qualitative and based on the study of procedures (both formal and informal) as they can be derived from constitutions and various laws, international treaties, guidelines and reports on the functioning of independent agencies and regulators. In Table 5.1 we range such methods under aggregate even though one probably cannot rely on quantitative information. It  is nonetheless based on data. The method here goes in the direction of process-tracing and qualitative casestudy analysis, and relies on documents and interviews. However, one also finds attempts at quantification, as in the case of indices such as those of central banks’ independence depicted in Figure 5.3. Crowe and Meade (2007) compiled a dataset that collected central bank indicators across OECD (Organisation for Economic Co-operation and Development) nations, and found a sharp increase in the protection of central banks from political influences in most cases. In the case of EU member states, previous attempts to measure the “limitations” placed upon individual governments have seen the calculation of a Banzhaf index based on each country’s number of votes in the Council under qualified majority rule, but newer instruments will need to be developed especially following the power shifts that occurred during the Eurocrisis. Debates surrounding the effective “neutrality” of central banks and the genuine “depoliticization” of policy at the European level are already reviewing the degree of technocracy present at the systemic level (Adolph 2013, Sánchez-Cuenca 2017).

106  Eri Bertsou and Daniele Caramani 1.00 0.80 0.60 0.40 0.20

Australia Austria Belgium Canada Chile Czech Republic Denmark Estonia Finland France Germany Greece Hungary Iceland Ireland Israel Italy Japan Korea Latvia Lithuania Luxembourg Malta Mexico Netherlands New Zealand Norway Poland Portugal Slovak Republic Slovenia Spain Sweden Switzerland Turkey United Kingdom United States

0.00

1980–1989

2003

Figure 5.3  Central bank independence: Index of Central Bank Independence in OECD (Organisation for Economic Co-operation and Development) countries. (Based on data from Crowe, C. and Meade, E.E., J. Econ. Perspect., 21, 69–90, 2007.)

Conclusion Building on a theoretical understanding of technocracy and its relation to democracy, the aim of this chapter was to advance the discussion to the operationalization and measurement of technocratic elements within democratic systems. Only then will we be able to study technocracy in a comparative perspective and answer pressing questions regarding its prominence, increase and decrease, causes and consequences. The process of depoliticization and the expanding roles of non-partisan experts in the ranks of parties, cabinets and beyond has been the focus of scholarly work in separate fields of research, without necessarily referring to technocracy. Our aim was to draw these together, under one coherent conceptualization of technocracy, and offer a map for levels of analysis and types of empirical measurement that can facilitate comparative research. We find that measuring technocracy is closely related to measures of depoliticization intended as the negative view of diverse interests and values, the competition for their allocation and alternative proposals for their aggregation (the core of party democracy). Technocracy is the desire to take the “politics” out of the policy-making process, and it therefore correlates with low levels of trust towards actors primarily pursuing their own interest instead of aiming for the good of a putative “whole”. Yet again, the degree to which preferences for technocracy (in citizens or given ideologies or social groups) are correlated with low levels of trust is an empirical question that makes sense theoretically but still needs to be verified through empirical data in a comparative and longitudinal manner.

Measuring technocracy  107 Technocracy is present in a pervasive way in a political system. It  is part of the preferences and attitudes of individuals (what behavioural analysis can reveal) in a way more or less correlated with individual socioeconomic features such as education and profession. Individuals with technocratic preferences and socioeconomic “predispositions” may vote for parties that present discourses favourable to technocratic governance, or at least are less critical of it. They may be more favourably oriented towards supranational integration and international agreements that take problem solving away from electoral competition and party politics. In countries with electorates thus oriented, the appointment of technocratic cabinets may be less controversial—or not. Following the conceptualization provided in the Introduction, technocracy is a matter of degrees. Establishing how much and at what levels technocracy exists across countries and over time is a matter of empirical investigation and, therefore, in need of solid indicators. Also, establishing what the configuration of technocracies is among the different parts of the political system should not be left to theoretical speculation, but rather to empirical verification. The  following chapters of this volume do precisely that, and offer insightful studies of technocratic politics in practice. While Chapters  6 through 8 deal with technocratic executives (based on sociological/compositional analysis) and their role in the policy process (both nationally and internationally), Chapter 9 investigates the framing of a referendum campaign in technocratic (vs populist) terms from a discourse-analysis perspective. Chapters 11 and 13 are based on “demand-side” individual data looking at attitudes. Chapters 10 and 12 approach technocracy from a systemic perspective, in Europe and in Latin America. Parts II and III of the volume are therefore concrete attempts to apply the concepts and methods developed in Part I.

Notes 1 These are the Fischer government in the Czech Republic (2009–10), the Bajnai government in Hungary (2009–10), the Monti government in Italy (2011–13), the Papademos government in Greece (2011–12) and the Ciolos government in Romania (2015–17). See Marco Valbruzzi in Chapter  6 for more details on their remit and composition. 2 See Dargent (2015) and Fischer (2009). Bickerton and Invernizzi Accetti (2017) discuss technocracy from a critical analysis perspective in the wake of thinkers such as Laclau, Mouffe and Rosanvallon. The  relationship between technocracy and party representation and its commonalities and differences with populism are discussed in Caramani (2017). 3 We limit the discussion of the empirical measurement of technocracy to representative democracies even though many of the arguments presented can be extended to authoritarian systems of various sorts (on Latin America’s regimes, for example, see Centeno 1993, 1994, Centeno and Silva 1998, Dargent 2015 and de la Torre 2013). 4 Much of this discussion follows the contrast between responsive vs responsible representation first highlighted by Birch (1964) and Mair (2009) and discussed extensively in Bardi et  al. (2014a, 2014b), but also in much of the literature on populism (see Kriesi 2014 for a definition and a review).

108  Eri Bertsou and Daniele Caramani 5 This is another point of contrast with representative party-based democracy, where the best outcomes are believed to result from the competition and compromise between different societal groups. 6 At this point it is important to note that the scientific approach can potentially be misappropriated and proponents of certain policies or ideas may use “pseudo-scientific” methods to advance their views. 7 The role of technocratic governments attracted scholarly attention in the early 2010s with such prominent appointments in European democracies, and led to a first systematic classification of technocratic governments (Brunclík 2015, 2016, McDonnell and Valbruzzi 2014, Pastorella 2016). 8 Based on a definition of “what does a technocrat look like” (both in terms of sociological traits and attitudes), one can classify political parties and cabinets based on their composition, promotion of political figures, reliance on external expertise and so on. 9 An additional element that enters public debates on the classification of officials is the status of the educational institution they have attended. A focus on elite schools is in line with the elitist dimension of technocracy, but need not be a criterion in this case. 10 This leaves open the question of whether once an official with expertise and experience in technical matters of governance enters the electoral competition and operates within a political party, they cease to be a technocrat. After the end of his technocratic cabinet in 2013, Mario Monti formed a political party and ran as its leader in the Italian elections. With 8.3% of the vote, it still lists 37 deputies and 19 senators, and was part of the grand coalition supporting a short-lived cabinet. Similarly, Emmanuel Macron, who served as a non-party aligned Finance Minister under President Hollande, created his own party and was elected president of the French Republic in 2017, appointing a large number of people without previous political experience, but with expertise in their fields. The  commitment to pursuing necessary reforms and welcoming people with expertise in the shaping of policy has earned him the title of technocrat in the press. 11 The coding exercise identified the speaker or author of the text as well as variables such as date, type of text, length, audience, language, source and contextual information. Grading ranged from −2 to +2 (opposite high intensity, opposite low intensity, absence, low intensity, high intensity of technocratic discourse). The coding unit was the paragraph or thematic-based sections in speeches where paragraphs could not be separated. Across paragraphs values were averaged, whereas across the five dimensions values were added together. 12 The  project, led by the Department of Media and Communication, has analysed media and political discourses as in quality press, tabloids, free newspapers, television debates, tweets, Facebook posts, and party manifestos in 12 European countries. The  sampling has created a text base of roughly 55,000 texts coded by 87 trained coders through the tool Angrist and following a pilot to test the inter-rater reliability. See Wirth et al. (2016) for detailed information. 13 Potential drawbacks of this operationalization include the degree of transparency politicians are prepared to allow regarding their representation role and their willingness to go against their constituents, party and the people in general. Examples of a survey at MP level can be found in the PartyRep project (Deschouwer and Depaw 2014), in particular, question 11.4 on the desirability “to delegate more decision-making to experts and independent agencies”. 14 This  type of analysis would allow for cross-country comparisons and the identification of longitudinal trends over time in the attitudes towards, or preferences for, technocratic governance. 15 The World and European Values Survey includes, tellingly, a single question, namely: “Do you think having experts, not government, make decisions according to what

Measuring technocracy  109 they think is best for the country is a very bad/fairly bad/fairly good/very good way of governing this country?”, which was then used in other work to explain and trace the development of technocratic preferences. No relevant items on technocratic attitudes exist in the Eurobarometer or the Latinobarometer. 16 The survey was administered in June 2017, with around 45 questions in 9 European countries. Beside socio-economic indicators and the items measuring technocratic preferences and attitude, the questionnaire also includes items on populist attitudes formulated to allow a comparison with existing studies (Akkerman et al. 2014, 2017). 17 A  more recent wave of studies uses conjoint experiments embedded in surveys, which present respondents with options regarding policy initiatives, decision-making processes and the involvement of independent experts in politics. These help to gauge not only the appetite of citizens for expert-driven governance, but also their relative preferences over other modes of decision making and compared across policy areas (Strebel et al. 2019, Bertsou and Caramani 2017). 18 Understandably, but also limiting, most work is on executives and focuses on their composition: e.g. the number of cabinet posts (and importance) given to technocrats, and in particular whether or not the prime minister is a technocrat or not. 19 For similar attempts on a variety of cases, see Brunclík (2015, 2016), Marangoni and Verzichelli (2015), Pastorella (2014, 2016), Costa Pinto et al. (2017). 20 This points to the responsible party model as the democratic vehicle through which democracy is enhanced. Democracy is unthinkably safe in terms of political parties (see APSA 1950, Sartori 1982, Schattschneider 1942). 21 See also the Introduction to the volume. Sánchez-Cuenca (Chapter  2) provides a useful distinction between two ways of depoliticization: directly, through the delegation of political decisions to independent agencies; and indirectly, through the establishment of strict rules that constrain elected officials to predetermined courses of action (see also Sánchez-Cuenca 2017). Further points on this issue can be found in Chapter 3 by Pier Domenico Tortola and in Chapter 4 by Reinout van der Veer. 22 The question here is really about the formal and practical relationship between independent bodies and political institutions, which is different than what has been seen in the institutional level of analysis—that is, the colonization of bureaucracies and non-elected bodies in general by politically affiliated and appointed personnel, i.e. the “partyness” of these bodies.

Part II

Institutions, actors and policies

6 Technocratic cabinets Marco Valbruzzi

Introduction: technocracy in the executive and its diffusion in contemporary Europe It  is customary to claim that technocracy is a complex, multidimensional and “essentially contested” concept (Gallie 1956). Accordingly, it is impossible for scholars concerned with the meaning and the evolution of technocracy in contemporary politics to agree on at least a minimal definition of the concept. This is especially true if one takes into consideration the variety of approaches that have dealt with this political phenomenon over time, from philosophy to organization studies, from sociology to the analysis of public policy, from political science to history. These different approaches and visions of technocracy emphasize a specific element or dimension of technocracy, but thus far the final outcome has not been a harmonious combination of all the distinct components. Certainly, the concept of technocracy cannot be analysed and treated in a simplistic manner. Nevertheless, this chapter aims to show that technocracy is neither a complex nor a multidimensional concept. Quite the contrary: at least for those scholars working in the field of political science, the concept of technocracy is a normal concept that can be analysed through the rules of traditional concept analysis. This  does not  mean that technocracy is an easy concept to deal with; it simply implies that we have at our disposal some useful guidelines for concept analysis that can be applied to the case at hand (Collier and Gerring 2008, Collier and Levistky 1997, Sartori 1984). Accordingly, the second section of the chapter will propose a solution for the analysis of the technocratic phenomenon in political science. In particular, it will show that technocracy as a concept can move in an orderly fashion along a revisited ladder of generality, whereby it is possible to define it by genus proximum et differentiam specificam, but also clearly specifying the layer of our analysis. This is important because different layers or slices of the concept of technocracy require different treatments.1 The third section will focus on a specific layer of the concept of technocracy— namely, the appearance of technocratic elements in the executive arena of contemporary political systems, in particular in the context of European parliamentary regimes. The discussion will not be on the nature and origin of technocracy, but

114  Marco Valbruzzi exclusively on the definition of real-existing cases of technocratic cabinets, such as those that have been formed throughout Europe in the last decade. Sections  “Technocrats and Technocrat-Led Cabinets” and “Composition of Technocrat-Led Governments” will provide a classification of these technocratled governments according to their partisan composition and remit. Then, on the basis of this classification, the few cases of what can be term “full technocratic”— as opposed to simply technocrat-led—governments will be presented and analysed. The last section will briefly discuss the possibility for future research based on the analyses and the conceptualizations/operationalizations of technocracy put forward in this chapter.

How to analyse the concept of technocracy The concept of technocracy is not, by definition, a fuzzy or essentially contested concept. Nevertheless, it can be described as a multilocus concept, in the sense that it can be applied to different loci—that is, different places, contexts or scenarios. In  light of this description, it is useful to analyse the concept of technocracy by making reference to the framework for concept analysis devised by Sartori (1984) and then further elaborated by Collier and Mahon (1993), whereby the analysis should be carried out, whenever possible, according to a hierarchical, taxonomic model based in principles of appropriate classification. Unfortunately, this is not what is happening in the current debate on the consequences of technocratic governments for either the stability or the quality of democratic regimes. Indeed, in the literature we can observe many cases of miscomparison—that is, comparisons of concepts located in different positions along the scale of generality. For  instance, scholars maintaining that technocracy is not compatible with democracy should specify that their argument can be applied only if both concepts are considered at their highest level of generality. Yet by moving the concept of technocracy to a lower level or including additional dimensions, the possibility of integration between technocracy and democracy becomes much more concrete. However, it should be clear that the movement along this dimensional scale follows precise, logical rules. More accurately, the ladder to universal or general concepts can be ascended by decreasing their connotation (i.e. the properties associated with a particular concept) and by increasing their denotation (i.e. the number of cases to which the concept can be applied). With some important caveats, this framework for analysis can be applied to the concept of technocracy. Figure  6.1 includes the tentative configuration of technocracy. At the more abstract level the concept of technocracy is considered as a type of political regime.2 This layer of the concept represents the primary category—or, in other words, a sort of technocratic ideal-type. More specifically, the form of government associated with the concept of technocracy can be considered as a subtype of the aristocratic regime, defined as that “form of political association in which the major decisions regarding policy are habitually made by a small section of the community acting through procedures and institutions

Technocratic cabinets  115 Technocracy as a form of government

Technocracy as a mode of governance

Technocracy as a type of cabinet

Technocracy as a set of ideas/attitudes

Figure 6.1  Graphical configurations of the concept of technocracy as a “radial” category

adapted to this end” (Merriam 1938: 857). The technocratic subtype adds a qualitative specification to this definition. Indeed, technocracy is not only the rule of the few but, more importantly, the rule of those few who master or control technical knowledge and expertise. Moreover, according to Charles Merriam, democracy and aristocracy are not compatible because they are based on distinct “modes of validation” (Merriam 1938: 857). In  other words, the validity of political authority in a democratic regime is based on the citizens’ consent, or, more broadly, on the consent of the public opinion (doxa). By contrast, the validity of authority in a technocratic regime derives from the knowledge (episteme) that a given group of experts is able to control. Arguably, at this level of generality technocracy is a non-democratic regime, and there is no possibility of integration with democracy. However, at a different level of generality that boundary between technocracy and democracy has to be examined in a different light. When technocracy is considered as a type of cabinet or, more precisely, a diminished subtype, then democratic regimes can accommodate or accept, to some extent and always with unavoidable tensions, the presence of technocrats in the executive arena. It  is important to stress here that by making explicit reference to the executive power, all those cases in which technocrats exercise their influence on the decision makers through their positions in the public bureaucracy are ruled out. That said, the concept of technocracy (qua technocratic cabinet) can be applied to different political regimes. In other words, technocracy as a type of cabinet can occur in a democratic as well as in a non-democratic setting. As will be specified

116  Marco Valbruzzi in the next section, a technocratic cabinet can coexist, under specific circumstances, with representative democracy. Therefore, at this level of comparison there is no a priori incompatibility between democracy and technocracy. Moving to a different slice of the concept under investigation, technocracy leaves the executive or governmental arena and turns into a particular mode of governance. Up to this point, technocracy has been defined and analysed by taking into consideration the nature and the mode of validation of political authorities. Now, by looking at technocracy as a mode of governance we are moving from the nature of the decision makers to the process (or locus) of decision making. If the former implied the existence of technocrats in the executive arena, the latter can be carried out regardless of whether the power-holders are technocrats or not. Hence, technocracy as a mode of governance describes a situation in which the process of decision making tends to be progressively depoliticized by enlarging the domain of impartial, value-neutral solutions and the influence of experts or non-political authorities over party politicians and partisan institutions. Following Urbinati (2014a: 230), this situation can be described as a sort of “epistemic democracy” aiming to “depoliticize democratic procedures and make them a method for achieving ‘correct outcomes’ or neutralizing partisan majority-based decisions”. It is debatable whether, or the extent to which, technocratic governance contradicts the principles sustaining democratic institutions. Especially in those circumstances characterized by multilevel arenas of decision making, a certain degree of technocratic governance can be accommodated in the domain of parliamentary democracies. Nevertheless, if technocracy as a mode of governance prevails over the functioning of democratic institutions, this “epistemic and unpolitical correction would change the opinion-based character of democracy and make it an expression of the power of knowledge” (Urbinati 2014a: 230). Lastly, the concept of technocracy can deal with politicians’ and citizens’ orientation towards politics and decision making. At this level, technocracy is neither a form of government nor a mode of governance, but it takes into consideration the general attitudes of political authorities or the general community. Many recent studies are good instances of what one can define as the ideational approach to technocracy. Despite their different methods, all these studies treat technocracy as a more or less coherent set of ideas, dealing with the functioning of the political process. This is not the place to establish whether technocracy is a (thin-centred) ideology, a political discourse, an attitude, a style, etc. For the purpose of this chapter, it is sufficient to acknowledge that technocracy can be analysed as a set of ideas, both from the supply side (parties and politicians) and the demand side (citizens and voters). Again, at this level, the concept of technocracy can coexist within a democratic setting, even though the diffusion of technocratic attitudes in a given society may be interpreted as a sign of progressive mistrust or discontent towards the working of democracy. Summing up, the concept of technocracy can be analysed at different levels of generality or taking into account different reference categories—and not all

Technocratic cabinets  117 Table 6.1  A typology of “technocratic categories” Opposition to democracy

Nature of decision makers

Experts (+) Non-experts (−)

Yes (+)

No (−)

Pure Technocracy (rule by experts) Technocratic governance

Technocratic cabinet Technocratic attitudes

conceptualizations imply a contradiction with the principles of democratic regimes. As shown in Table 6.1, technocracy is the opposite of democracy when defined at its highest level of generality—that is, as a system of government. In other circumstances, for instance when technocracy is considered as a specific type of cabinet or a set of ideas and attitudes (namely, as diminished subtypes of the primary category), its opposition to democratic principles is strongly mitigated. The  cohabitation between democratic principles, on the one hand, and technocratic practices or orientations, on the other, is never easy or inconsequential. Nevertheless, in acknowledging that under specific circumstances technocracy and democracy can coexist, it does not mean that they should be regarded as usual partners. Indeed, democracy and technocracy remain strange bedfellows, but their rendezvous have become more frequent than in the past.

Technocracy in the executive arena This section aims provide an ideal-type definition of a technocratic government,3 and it takes as starting point what ought to be the exact opposite of a technocratic government: namely, a party government (Caramani 2017). Here, unlike the eventual “technocratic government” destination, there exists a lot of rich political science literature (e.g. Blondel and Cotta 2000, Katz 1986, 1987, Mair 2008, Rose 1969). For the purposes of this chapter, it is useful to rely on the work of Richard Katz (1987: 7), who sets out five necessary conditions for an ideal-type model of party government: (1) Decisions are made by elected party officials or by those under their control. (2a) Policy is decided within parties which (2b) then act cohesively to enact it. (3a) Officials are recruited and (3b) held accountable through party. Like all ideal-types, this is, as Katz (1986: 42−43) says, an intellectual construct “whose logic is far more coherent than is the actual operation of any real government”. The passage from “ideal-type” to “real-existing party governments”

118  Marco Valbruzzi not only implies a distancing from the ideal, but also a move from singular to plural—i.e. from a single ideal-type to a plurality of types. As Jean Blondel (2000: 97) notes, “party government is often mentioned in the singular as if it was of one type only. In  reality, there are large variations among different types”. It is one thing, however, to show that there are many and varied forms of party governments; it is another to identify clearly when we can no longer speak of party governments and enter instead the area of technocratic governments. Blondel (1991: 14) raises the question of the differences between party and technocratic governments when he observes, in reference to the composition of the former, that “there has to be a limit, however, as the appointment of nonparliamentarians may strain the relationship of the government with parliament”. Unfortunately, he does not offer a roadmap for how and where to set this limit. Nor does Maurizio Cotta (2000a: 214), who asserts that “the ‘party-dependent government’ model is a polar extreme from which real-world situations are distant, albeit more or less”, yet gives no indication about the location of the point at which we are so far from this polar extreme that we are no longer dealing with a party government, but a technocratic one. Before moving on, it is worth acknowledging here that some might object that technocratic governments often do have a low degree of partisanship and, as such, can be placed towards the far end of a party-government continuum. Indeed, even in the cases of those technocratic governments which contain no party representatives, one could contend that these governments still depend on parliamentary support and, hence, ultimately on parties. As such, although located at the furthest point from the polar extreme of the party government ideal-type, technocratic cabinets can nonetheless be viewed as still operating within that pole’s magnetic field. Although these types of cabinet do remain accountable to parliaments, it would be detrimental to our understanding of each type to analyse technocratic governments as extreme cases of party government. In fact, technocratic governments can have profoundly different geneses, competences and, often, consequences than “normal” party governments. Including these two types of government in the same conceptual container does not allow us to identify their specific and distinct features. The main premise here is thus that a technocratic government is different from any form of party government. The next step is to identify its main characteristics. If we accept that a technocratic cabinet, in its ideal-type form, ought to be a mirror image of the ideal-type of party government, then it is possible to return to Katz’s conditions for party government (Katz 1986, 1987) and say that an ideal-type technocratic government should be one in which the following three necessary conditions are satisfied: 1. All major governmental decisions are not made by elected party officials; 2. policy is not decided within parties which then act cohesively to enact it; and 3. the highest officials (ministers, prime ministers) are not recruited through party.4

Technocratic cabinets  119 By working backwards, therefore, it is possible to construct an ideal-type model of a technocratic government, which is distinct from any model of party government. The  next step consists in moving from this ideal-type of technocratic government to those real-existing technocratic cabinets.

Technocrats and technocrat-led cabinets Just as there are many different types of party government, at varying distances from Katz’s ideal-type definition, so too may there be different types of techno­ cratic governments at varying distances from the ideal-type definition. Before attempting to classify these, however, the first task is to establish how many potential technocratic cabinets one is dealing with in Europe. It seems sensible to begin by identifying all those governments which have been led by a technocrat, before sifting through these, classifying them and establishing similarities and differences.5 This first task, however, presents another definitional question: how do we decide who is and who is not a technocrat? A  good starting point here is the observation by Jean Meynaud (1964: 262) that “when he becomes a technocrat, the expert becomes political”. This implies two things. First, that a technocrat is an expert of some description. Second, “the technocrat does politics” (1964: 259). This  latter point is important since commentators have on occasion misleadingly referred to technocrat-led governments as “apolitical” or “non-political”.6 With regard to the question of how we can identify technocrats, Cotta and Verzichelli (2002: 145) define a technocratic minister as one “totally lacking in both a parliamentary and party political background and having, rather, some specialist background that is related to the ministry he or she occupies”. While useful, this definition is too vague to be employed empirically since Cotta and Verzichelli do not explain precisely what a “party political background” consist of.7 To overcome this lack of clarity about the degree to which a technocrat should not be “party political”, the following operational definition is adopted: A prime minister or minister is a technocrat if, at the time of his/her appointment to government, he/she: (1) has never held public office under the banner of a political party; (2) is not a formal member of any party; (3) is said to possess recognized non-party political (i.e. technical) expertise which is directly relevant to the role occupied in government. Taking these three conditions as together necessary and sufficient, we now proceed to identify all those democratic cabinets in the EU-28 which have been led by a technocrat since the end of World War II. As we can see from Table 6.2, there have been 28 cases of technocrat-led governments up to December 2018 in 9 member states (Bulgaria, Croatia, Czech Republic, Finland, Greece, Hungary, Italy, Portugal and Romania). Of these 28, 19 also had a technocrat Minister of

120  Marco Valbruzzi Table 6.2  Technocrat-led governments in EU-28 member states, 1945–2018 Member state Bulgaria

No. Prime Minister

3 Berov Indzhova Raykov Croatia 1 Orešković Czech 3 Tošovský Fischer Republic Rusnok Finland 3 Von Fieandt Lehto Liinamaa Greece 5 Grivas Zolotas I Zolotas II Papademos Pikrammenos Hungary 1 Bajnai Italy 4 Ciampi Dini Monti Conte Portugal 2 Nobre da Costa Pintassilgo Romania 6 Stolojan Vacaroiu I Vacaroiu II Vacaroiu III Isărescu Cioloș Total 28

Period in office

Days Technocrat % of in minister of technocratic office economy ministers

30/12/1992–17/10/1994 18/10/1994–25/01/1995 13/03/2013–29/05/2013 22/01/2016–19/10/2016 01/02/1998–17/07/1998 08/05/2009–13/07/2010 10/07/2013–25/10/2013 29/11/1957–18/04/1958 18/12/1963–12/09/1964 13/06/1975–21/09/1975 12/10/1989–23/11/1989 23/11/1989–13/02/1990 13/02/1990–11/04/1990 11/11/2011–16/05/2012 16/05/2012–17/06/2012 15/04/2009–14/05/2010 29/04/1993–10/05/1994 17/01/1995–17/05/1996 16/11/2011–27/04/2013 01/06/2018– 29/08/1978–15/09/1978

656 99 77 270 166 431 107 140 269 100 40 82 57 187 31 395 376 486 528 — 18

Yes Yes Yes No No Yes Yes Yes Yes Yes No No No No Yes Yes No Yes Yes Yes Yes

80 100 100 28 38 100 87 100 100 100 76 32 68 17 100 50 44 100 100 35 100

31/07/1979–12/12/1979 16/10/1991–27/09/1992 13/11/1992–18/08/1994 19/08/1994–01/09/1996 02/09/1996–03/11/1996 21/11/1999–26/11/2000 10/11/2015–11/12/2016

147 347 644 745 101 373 335

Yes No Yes Yes Yes No Yes

100 10 50 32 41 5 100

Sources: Bell, J.D., Democratization and political participation in postcommunist Bulgaria, in Dawisha, K. and Parrott, B. (eds.), Politics, Power and the Struggle for Democracy in South-East Europe, 353–402, Cambridge University Press, Cambridge, 1997; Conrad, C.R. and Golder, S.N., Eur. J. Polit. Res., 49, 119–50, 2010; Protsyk, O., East Eur. Polit. Soc., 19, 135–60, 2005a; Protsyk, O., Eur. J. Polit. Res., 44, 721–48, 2005b; Sonntag, L., Politica, Available at www.kolumbus.fi/taglarsson/, 2010; Strøm, K., Müller, W.C. and Bergman, T., Delegation and Accountability in Parliamentary Democracies, Oxford University Press, Oxford, 2003. Notes: Data for Greece, Portugal and Spain have been collected after their democratization in the 1970s. Similarly, in the case of new EU member states from Central and Eastern Europe, we consider only the post-1989 period.

the Economy (which have been included as a further indicator of the extent to which key cabinet posts were occupied by non-party ministers).

Composition of technocrat-led governments While the previous section listed those cabinets in Europe which have been led by technocrats, it should also be made clear that all such governments are not  necessarily fully technocratic. In  other words, some technocrat-led

Technocratic cabinets  121 governments are more technocratic than others. This section and the next aim at demonstrating how and why this is the case. From even a brief glance at Table  6.2, one can see significant variation between real-existing technocrat-led governments as regards both duration and composition. These are important differences, but they unfortunately tend to be overlooked in discussions. In particular, the mere presence of a technocrat at the head of a government is often taken as automatically rendering that cabinet equally as technocratic as all other technocrat-led ones. The “lumping together” by many commentators of the governments which took office in Greece and Italy in November 2011 is a good example of this flawed logic, with the likes of Vivien Schmidt (2011) and Andrew Moravcsik (2012) treating the administrations led by Papademos in Greece and Monti in Italy as technocratic and equivalent. This, however, ignores the considerable differences between the two. In Greece, Papademos became prime minister of a cabinet composed mostly of representatives from the two largest parties at that time: PASOK (Panhellenic Socialist Movement) and New Democracy. In  fact, 12 of his 18-member cabinet had been PASOK ministers in the previous George Papandreou government and, as we can see from Table 6.2, only 17% of the cabinet was composed of technocrats. The Papademos government can thus be seen as a temporary, oversized coalition which was led by a technocrat (in addition to the two main parties, it also contained the minor party Laos for the first three months). By contrast, in Italy, the outcome of the fall of Silvio Berlusconi’s government was very different to that of George Papandreou’s in Greece: the government led by Monti did not contain a single party representative and was nominated with the stated intention of lasting well over a year in office and bringing in major reforms (Bosco and McDonnell 2013, Pasquino and Valbruzzi 2012). The  same holds for many other governments in Table  6.2, including several which occur in the same country. To take another Greek example, the first Xenophon Zolotas administration in 1989 was akin to the Papademos one in 2011 since it was formed by the three main parties and included party politicians and technocrats (who, in the case of Zolotas I, were nominated by the parties and, in that of Papademos, were considered linked to the parties) (Dinas and Rori 2013). Likewise, in terms of their remits, both governments were charged with taking urgent steps over a six-month period to tackle an economic crisis (Pridham and Verney 1991). By contrast, the second Zolotas government in 1990 appears much more similar to that led by Ioannis Grivas in 1989 (indeed, the two cabinets had more individual ministers in common than was the case for Zolotas I and II) and to the 2012 Panagiotis Pikrammenos administration. All three governments (Grivas, Zolotas II and Pikrammenos) contained a majority of technocrat ministers and were what the Greeks call “service governments”— i.e. they deal only with the day-to-day business of government until new elections are held within a very short space of time. A final case worth mentioning here, given its borderline status8 in terms of partisan/technocrat composition, is the government led by Gordon Bajnai in Hungary from April 2009 to May

122  Marco Valbruzzi 2010. On the one hand, Bajnai was not a sitting member of parliament when he became prime minister, he and half his cabinet were officially “independent” (with many ministers coming from outside politics), and his government was pitched as a “crisis-management” administration (Várnagy 2010: 1003−4). On the other hand, Bajnai had been a minister in the previous Ferenc Gyurcsány MSZP (Hungarian Socialist Party) Socialist government, the non-independent half of his cabinet was composed solely of MSZP elected representatives, and it was supported in parliament only by the MSZP. As a result, according to country experts, it was apparently treated by the opposition as an MSZP government. So, while Bajnai fulfils the three criteria for definition as a technocrat and his government can therefore be included among the cases of technocrat-led governments in Europe, it seems clear that this case is rather far from our ideal-type definition of a technocratic government.

What can technocrat-led governments do? While extremely useful in understanding the differences between technocratled governments, partisan composition is not  the only benchmark by which these governments should be divided and classified. As we saw in Table 6.2, in addition to the large variations in terms of partisan composition, there are also significant differences in duration. The issue of how long a technocratled government lasts is usually linked to the broader question of its remit. In other words: what can these governments do? As Cotta (2000b: 91) says, government is not just about those “who form part of it at the different levels: it is also constituted by the activities it undertakes, by the policies it decides”. Taking a pair of Greek and Italian cases as examples, if one looks only at party composition, the Dini and Pikrammenos governments appear identical since they were composed entirely of technocrat ministers. However, it would be foolhardy to ignore the fact that the Pikrammenos government lasted just 31 days and was put in place simply to tend to basic daily administration until new elections could be held, while the Dini government lasted more than a year and introduced major reforms. Thus, remit (i.e. what governments can do) counts as well as composition. Considering remit, however, means tackling another problematic category: the caretaker government. While there are many definitions of this term in the literature, unfortunately one again finds confusion. Petra Schleiter and Edward Morgan-Jones (2009: 674) state that “caretaker situations arise when governments resign pending early elections, following, for example, coalition failures or parliamentary dissolution” (for a similar view, see Van Roozendal 1997: 72). The  problem with this is that it excludes those caretaker governments which arise after a general election when the parties are unable to form a government. Leaving aside the issue of when caretaker governments occur, most definitions do concur on two key points. First, caretakers are said to exert a “bridging role”

Technocratic cabinets  123 between duly-mandated governments for a brief (but unspecified) period of time. Second, it is broadly agreed that caretakers cannot undertake the same range of policy actions as a duly-mandated party government. As Valentine Herman and John Pope (1973: 196) note, “it is accepted that these governments have a limited life-span and limited freedom of action”. This acceptance of a restricted remit for caretakers is reflected in national political cultures. As Sona Golder (2010: 4) points out, most countries “have a strong norm that caretaker governments … do not  have the authority to make major policy initiatives”. This  acknowledgement may be either formal or informal. For example, in Greece, constitutional law requires the caretaker cabinet to be “as far as possible, of wide acceptance in order to conduct the elections” (Koutsoukis 1994: 280). Similarly, in Belgium, the head of state can ask a government which has resigned “to remain as a caretaker and attend ‘to current affairs’ (but no politically important matters) until a new government can be sworn in” (De Winter et al. 2000: 343). Even in countries such as Norway and Sweden, where the constitutions do not  contain any particular rules or constraints regarding caretaker governments, “in reality, caretaker cabinets have always avoided presenting bills to Parliament and have confined themselves to administrative matters” (Larsson 1994: 172). To return to the question of caretakers and technocrats: not  all caretaker prime ministers are technocrats. One can find caretaker governments led by party representatives in countries such as the Netherlands and Belgium which have no history of technocrat-led governments while, in countries like Finland (Raunio 2004, Törnudd 1969) and Italy, it is possible to observe both party-representative-led caretaker governments and technocrat-led caretaker governments. However, based on the above discussion of what caretaker governments can do, we also know that not all technocrats are caretakers. By looking again at the cases in Table 6.2, it can be said that those led by Indzhova in Bulgaria and Tošovský in the Czech Republic were caretakers, while those led by Berov in Bulgaria and Fischer in the Czech Republic were not: Indzhova and Tošovský could only concern themselves with “minding the shop temporarily”, as Jaap Woldendorp, Hans Keman and Ian Budge (2000: 18) put it. However, this is not the case for Berov, Fischer and the other technocrat-led governments listed in Table  6.3, such as Zolotas I, Dini, or Vacaroiu I, II and III in Romania. If, as Golder states, caretaker governments “should simply maintain the status quo”, then the technocrat-led governments in Table 6.3 cannot be considered caretakers given both the major policy changes which they were charged with introducing and the length of time they were given to do so (Golder 2010: 4, see also Conrad and Golder 2010, Laver and Shepsle 1994). Differentiating between the list of technocrat-led administrations in this way—in terms of remit—thus allows us to move towards identifying, in the next section, what have been termed “full technocratic governments” (McDonnell and Valbruzzi 2014).

Table 6.3  Non-caretaker technocrat-led governments and their policies Government Period Zolotas I (Greece) Stolojan (Romania) Vacaroiu I, II and III (Romania) Berov (Bulgaria)

Main economic policies adopted or implemented

1989–1990 Reorganization of collective bargaining framework; VAT increase 1991–1992 Price liberalization; wage stabilization; privatization of some state-owned enterprises; decrease of restrictions on exports 1992–1996 Price and trade liberalizations; cuts in subsidies to state enterprises; privatization policies; introduction of VAT 1992–1994 Deficit reduction; privatization law; law on agricultural land ownership and use; introduction of VAT

Ciampi (Italy)

1993–1994 Reform of the collective bargaining system; public administration reforms; wage controls; adoption of a legal framework for supplementary pensions; electoral system reform Dini (Italy) 1995–1996 Introduction of a flexible retirement age; shift to a contribution-related formula for pensions; indexation of pensions to real wage growth; deficit cuts Isărescu 1999–2000 Introduction of new pension systems; financial sector reforms; (Romania) privatization law Bajnai 2009–2010 Income tax reductions; abolition of different special taxes; (Hungary) introduction of new property taxes; indexing of pensions to GDP; increase in the retirement age Fischer 2009–2010 Adoption of “anti-crisis package” (deficit cuts; increase in (Czech VAT and consumer tax; reduction of some state support Rep) benefits; suspension of plans to raise pensions; reduction of salaries for state employees) Papademos 2011–2012 Adoption of “anti-crisis package” (increase in tax properties; (Greece) introduction of special tax on high pensions; reform of pension system; cuts on public sector allowances) Monti 2011–2013 Massive reduction in the budget deficit; introduction of (Italy) property taxes; increase in the retirement age; cuts in the size and costs of administrative bodies; “liberalization” of closed professions, labour market reform Cioloș 2015–2016 Reform of the public bureaucracy (local and central (Romania) administration); adoption of anti-corruption policies Orešković 2015–2016 Reduction in the budget deficit (Croatia) Conte (Italy) 2018– Reform of the pension system (lowering the retirement age); introduction of the basic income; decree-law on security and migration; adoption of new regulations on temporary work Sources: Blokker, P., Continuity in change: Social consequences of economic reform in Romania, in Jilberto, A.E.F. and Riethof, M. (eds.), Labour Relations in Development, 95–120, Routledge, New York and London, 2002; Ferrera, M., West Eur. Polit., 20, 231–49, 1997; Ganev, G.Y. and Wyzan, M.L., Bulgaria: Macroeconomic and political-economic implications of stabilisation under a currency board arrangement, in Lundahl, M. and Wyzan, M.L. (eds.), The Political Economy of Reform Failure, 170– 96, Routledge, Abingdon, 2005; Hopkin, J., Int. Spect., 47, 35–48, 2012; Linek, L. and Lacina, T., Eur. J. Polit. Res., 49, 939–46, 2010; Pop, L., Democratising Capitalism? The Political Economy of PostCommunist Transformations in Romania, 1989–2001, Manchester University Press, Manchester, 2006; Roper, S.D., Romania. The  Unfinished Revolution, Routledge, Abingdon, 2000; Schmidt, V.A. and Gualmini, E., Comp. Eur. Polit., 11, 360–82, 2013; Várnagy, R., Eur. J. Polit. Res., 49, 1001−08, 2010. Abbreviations:  GDP = Gross Domestic Product; VAT = Value-Added Tax.

Technocratic cabinets  125

Real-existing full technocratic governments In light of the earlier discussion, it is possible to devise a typology of technocratled governments in terms of their composition and remit in order to arrive at what can be defined as real-existing “full technocratic governments”. The first criterion—partisan/technocrat composition—requires distinguishing between technocrat-led cabinets consisting mostly of technocrats and those in which party representatives make up the majority of ministers. There are, of course, borderline cases. The Bajnai government, whose composition has been discussed above, is  one such  case. The  first government led by Nicolae Vacaroiu in Romania between 1992 and 1994 is another, since it too was split 50/50 between technocrat and party ministers. In these “tiebreak” cases, the solution adopted when classifying them is to check whether the Minister of the Economy (or Finance, as it may be called) is also a technocrat. In other words, if both the premiership and what is generally considered the most important ministry (Economy/Finance) are occupied by technocrats, then it is possible to judge evenly split cabinets to be more technocratic than partisan. The  second criterion—remit—refers to the mandate of technocrat-led governments to change the status quo (i.e. to go beyond a caretaker role). Whether they do so or not  is another matter: it is of course possible that a government may not. But here we are interested in whether they have it within their power to do so. Caretaker governments take office with the knowledge that they have been created to look after daily administration, to “mind the shop” rather than to make major changes. However, this limitation does not apply to all technocratled governments. By combining these two criteria of composition and remit, it is possible to come up with the typology of technocrat-led governments shown in Table 6.4. The first type (in the bottom right-hand corner) is the technocrat-led partisan government, characterized by a majority of party representatives under a technocrat prime minister and with a mandate to change the status quo. There have been nine cases of this in the EU-28: Orešković in Croatia, Zolotas I and Papademos in Greece; Ciampi and Conte in Italy; and Stolojan, Vacarioiu II and III, and Isărescu in Romania. The second type (in the top right-hand corner) is that of the partisan caretaker, which contains a majority of party representatives in cabinet Table 6.4  A typology of technocrat-led governments Composition

Remit

Must maintain status quo Mandate to change status quo

Majority technocrat

Majority party rep

Non-partisan Caretaker Full Technocratic Government

Partisan Caretaker (Technocrat-led) Partisan Government

126  Marco Valbruzzi under a technocrat prime minister, but does not have a mandate to change the status quo. There have been no cases of this in the EU-28, but there is no theoretical reason why such a government could not  exist. The  third type (in the top left-hand corner) is the non-partisan caretaker. This  is characterized by a majority of technocrats in cabinet under a technocrat prime minister and does not have a mandate to change the status quo. There have been 12 such governments in the EU-28: Indzhova and Raykov in Bulgaria; Tošovský and Rusnok in the Czech Republic; Grivas, Zolotas II and Pikrammenos in Greece; Nobre da Costa and Pintassilgo in Portugal; and three cases in Finland (Von Fieandt, Lehto and Liinamaa). The fourth and final type (in the bottom-left corner) is what can be termed full technocratic government. These governments contain a majority of technocrats in cabinet under a technocrat prime minister and have a mandate to change the status quo. To date, there have been seven such cases in the EU-28: Dini and Monti in Italy; Bajnai in Hungary; Vacaroiu I and Cioloș in Romania; Berov in Bulgaria; and Fischer in the Czech Republic. These are the only governments that can be considered as real-existing full technocratic governments. In other words, they are technocrat-led governments whose composition and remit are such that they are qualitatively distinct from other types of technocrat-led administrations. All seven, although they do differ from one another in some respects, fulfil the following three necessary and sufficient conditions: 1. The prime minister is a technocrat; 2. the majority of ministers are technocrats; and 3. they have a mandate to change the status quo. Having presented this typology, one can now return to the list of cases presented in Table 6.2 and classify all technocrat-led governments in the EU-28 accordingly (see Table 6.5). It should be noted at this point that a technocrat-led government’s classification is not necessarily set in stone over time, and that the same government may shift from one of our four types to another due to changing circumstances. A  good example of this is Fischer’s government in the Czech Republic, which was created as “a caretaker government of experts” (Stegmaier and Vlachová 2011: 238). Its original remit was simply to “mind the shop” until a constitutional amendment could be passed allowing a general election to be called. However, after Fischer had been in office for a month, the Czech Constitutional Court rejected the amendment and so his government found itself continuing longer than had been envisaged. In  the meantime, the Czech Republic was faced with a severe economic crisis which required the adoption of strong austerity measures. As a result, the Fischer administration went from a non-partisan caretaker to a full technocratic government for most of its time in office (Linek and Lacina 2010: 939).9 It  is worth emphasizing, finally, that the other types of real-existing technocrat-led governments—non-partisan caretaker and technocrat-led partisan government—by far outnumber the full technocratic governments listed

Technocratic cabinets  127 Table 6.5  Types of technocrat-led governments in EU-28 member states, 1945–2018 Member state

No.

Prime Minister

Type of government

Bulgaria

3

Croatia Czech Republic

1 3

Finland

3

Greece

5

Hungary Italy

1 4

Portugal

2

Romania

6

Berov Indzhova Raykov Orešković Tošovský Fischer Rusnok Von Fieandt Lehto Liinamaa Grivas Zolotas I Zolotas II Papademos Pikrammenos Bajnai Ciampi Dini Monti Conte Nobre da Costa Pintassilgo Stolojan Vacaroiu I Vacaroiu II Vacaroiu III Isărescu Cioloș

Full technocratic government Non-partisan caretaker Non-partisan caretaker Technocrat-led partisan government Non-partisan caretaker Full technocratic government Non-partisan caretaker Non-partisan caretaker Non-partisan caretaker Non-partisan caretaker Non-partisan caretaker Technocrat-led partisan government Non-partisan caretaker Technocrat-led partisan government Non-partisan caretaker Full technocratic government Technocrat-led partisan government Full technocratic government Full technocratic government Technocrat-led partisan government Non-partisan caretaker Non-partisan caretaker Technocrat-led partisan government Full technocratic government Technocrat-led partisan government Technocrat-led partisan government Technocrat-led partisan government Full technocratic government

Total

28

in Table 6.5. This ought to encourage us to reflect on how we talk about technocratic governments, in particular in respect of their prevalence and significance in contemporary Europe. Put simply, there has not been the technocratic government invasion that many commentators would have had us believe. European governments remain, overwhelming and especially in West European countries, duly-mandated party governments.

Conclusion The topic of technocrats and technocratic government has hitherto remained a largely unexplored field, and this contribution aims primarily to establish some conceptual clarity which can also serve to provide more solid foundations for further research. On this last point, there are three main areas which scholars might profitably focus on. The first is to consider the conditions under which the governments discussed in this chapter occur. An initial observation, based on the data presented here, is that we seem to find them in political systems where the  head of state plays a decisive role in government formation (Blondel and

128  Marco Valbruzzi Müller-Rommel 2001, Brunclík 2015, Costa Pinto et al. 2017). A second observation is that these governments appear more common in states where the party system is either crumbling (such as Italy and Greece) or has not been fully rooted (the obvious examples are countries which have recently undergone a transition to democracy, such as Bulgaria and Romania in the early 1990s). These two variables—party system consolidation and the role of the head of state— thus seem useful starting points for additional work on the genesis diffusion of technocratic ministers across political regimes. The second main area for further research concerns policy. For example, comparative case studies could examine in far greater detail what such governments actually do in policy terms. They could also consider the real degree of policy autonomy of these governments from party groups and relevant external, especially supranational actors. This chapter has focused on what, following Richard Katz (1987), one might define as “the technocraticness of government”—namely, the presence and diffusion of non-partisan experts in the cabinet.10 However, “technocratic governmentness” (i.e. the spread of the role of experts in various areas of policymaking) is also an important area of research and should be related to that conceptualization of technocracy that references a specific mode of governance. On this point, it is worth noting that many party governments contain non-partisan experts in key ministries. And even those that do not may pursue policies which are designed and closely monitored by external technocrats. In  any case, the diffusion of technocratic governments has been accompanied and, to some extent, prompted by the increasing frequency of non-partisan experts in the executive. As shown in Figure 6.2, the level of technocraticness of

Technocraticness of government (%)

12

10

8

6

4 1945

1955

1965

1975

1985

1995

2005

2015

Figure 6.2  Technocraticness of government in EU-28 member states (%), 1945–2018

Technocratic cabinets  129 government in the EU-28 member states—that is, the proportion of technocrats over the whole population of ministers—has steadily increased during the postwar period. Overall, more than 10% of ministers in the last decade have been selected and nominated for their expertise rather than their partisanship, and this is true especially for Southern and Central European countries. All these different trends, in terms of both technocraticness of government and technocratic governmentness, lead us back to the crucial questions about democratic accountability mentioned above and which constitute the third main area of potentially fruitful research on this topic. Certainly, the issues concerning the rising frequency of technocrats in office are far too big an area to explore in-depth here, but the discussions and typologies presented throughout this chapter can serve as a useful analytical framework for this type of intellectual endeavour. Be that as it may, and pending other, more convincing arguments, for the moment the conclusion reached by Sartori (1982: 431) remains valid: “we find ourselves uneasily, and often lazily, placed between these two extremes: a government of non-experts over experts, or a government planned out by experts without democracy. If democracy is to survive, it will have to steer clear of either extreme”. My impression, corroborated by some evidence, is that our democracies are confusingly getting closer to those extremes, but the path in front of us is not completely laid out and the future remains open.

Notes 1 For  the analysis of these layers in the concept of technocracy and their respective operationalization, see Bertsou and Caramani, Chapter 5 in this volume. 2 As presented in this chapter, the concept of technocracy can be interpreted as a “radial” rather than a classical category (Collier and Mahon 1993). Yet this specification does not mean that the logic of concept analysis should be relaxed or abandoned; as Collier and Mahon (1993: 848) put it, “this can be accomplished by applying some form of multidimensional scaling that specifies underlying dimensions for comparing cases”. 3 This  section reproduces the conceptual analysis developed by McDonnell and Valbruzzi (2014). 4 It is worth noting that Katz’s criteria have been slightly adapted in order to fit technocratic governments in democracies: although the principal–agent relationship is a point of contention for such governments, it does not make sense to set as a condition that technocrat ministers and prime ministers are in no way under party control. Hence, it has been omitted the second part of Katz’s first condition. Likewise, in the third condition, it would be illogical to posit (at least in European democracies) that technocrat ministers and prime ministers are in no way accountable through party (point 3b in Katz’s criteria). 5 While it could be objected that this excludes governments made up of technocrats, but led by a party representative, it was decided not to consider these, for three main reasons: (1) the fundamental role played by the prime minister in parliamentary government systems (Poguntke and Webb 2005); (2) a technocratic government is defined by the distance which separates it from a party government—that is, one in which party representatives occupy the key roles, first and foremost that of prime minister. Hence, it seems contradictory to include a government led by a party representative within

130  Marco Valbruzzi the realm of technocratic governments; (3) to date in Europe, there has not been a single case of a government in which a party representative has acted as prime minister in a cabinet composed of a majority of technocrat ministers. 6 For  example, in the wake of Mario Monti’s appointment in Italy, Nathan Gardels (2012: 30) referred to “the depoliticized democracy being practiced by Prime Minister Monti”. 7 In a more recent contribution, Cotta (2018: 272) defines non-partisan ministers as “all those who previously to their first selection as minister had not held a national or local leadership position in the party and who had not been members of parliament (except if elected just before their ministerial nomination). They came to the cabinet from outside the world of party politics and, presumably, on the basis of different qualifications (typically, but not  exclusively, a specialized expertise in some domain)”. Despite its many exceptions and limitations, this definition is good for the analysis of non-partisan ministers, as a general category. But, by virtue of their expertise, technocratic ministers are a specific kind of non-partisan politician. Unfortunately, this aspect remains residual in the general definition put forward by Cotta. 8 Another borderline case is represented by the cabinet recently formed in Italy by the Five-Star Movement and the League (Valbruzzi 2018), under the premiership of Giuseppe Conte, a lawyer and professor of private law. Although the description of Conte meets all three established criteria for defining a technocrat, his case is nevertheless ambiguous because he was presented by the Five-Star Movement, before the 2018 general election, as a member of the cabinet in case of victory. 9 On the specificities of the Fischer’s cabinet, see Brunclík (2016) and Hanley (2018). 10 Technocraticness in government can be measured as the proportion of non-partisan ministers out of the total number of ministers included in a single cabinet.

7 Technocrats in cabinets and their policy effects Despina Alexiadou

Introduction Since the start of the financial and banking crisis in Europe, the number of nonelected, expert ministers, also known as technocrats, has more than doubled (as seen in Chapter 6; see also Alexiadou and Gunaydin 2019). Yet, we still have little knowledge of whether these appointments matter for policy outcomes (the topic of Chapter  10). This  is somewhat puzzling, particularly when it comes to the appointment of technocrat finance ministers who are responsible for a range of policies, from the government’s budget to tax rates. While there is evidence that ministerial appointments can, under certain conditions, affect policy outcomes (Alexiadou 2015, Chwieroth 2007, Jochimsen and Thomasius 2014), it is unclear whether the appointment of non-elected experts has independent policy effects. The  empirical evidence on the role of technocrat ministers as policymakers is limited. Researchers have primarily studied technocrat ministers within the context of major economic crises or political and economic transitions in select countries in Latin America (Domínguez 1997, Joignant 2011) and Eastern Europe (Greskovits 2001). Dargent’s and Kaplan’s recent contributions advance the systematic study of outside experts in government. Dargent (2015) shows that outside experts appointed to government departments due to their expertise have the power to set a policy agenda, even if it runs against the premier’s policy priorities, while Kaplan (2017) finds that economists hired during economic crises as a way to increase a government’s credibility are associated with fiscal contractions. While these accounts offer unique insights into the role of outside experts in government, they do not  systematically examine the policy role of technocrat ministers as opposed to expert politicians. This is due to the fact that existing works do not use a common definition of “technocrat”. For some, especially scholars who study Latin American countries, technocrats are all experts in government, irrespective of their political experience (Kaplan 2017, Santiso 2003), while for others, especially scholars who study Europe, technocrats are non-elected experts (as defined in Chapter  6). Unless a common definition of “technocrat” is adopted, it is not  possible to thoroughly examine their role in policy (for a detailed discussion see Alexiadou (2018), and also the Introduction and Chapter 5 in this volume).

132  Despina Alexiadou This chapter tests the policy effects of the technocrat and expert ministers of finance and social welfare, defined as non-elected experts appointed to cabinet positions (Alexiadou and Gunaydin 2019, Costa Pinto et  al. 2017, McDonnell and Valbruzzi 2014). More specifically, technocrats are defined as professionals, appointed to the cabinet in a ministerial post, with policy expertise within their department’s policy jurisdiction and who have never held elective office, either at the national, subnational or local levels (Alexiadou and Gunaydin 2019). According to this definition, expertise alone does not qualify a minister as a technocrat, nor does “independence” from politics. A technocrat is both an expert within her department and has a professional career outside politics (for a longer discussion on the definition of technocrat policymakers, see the Introduction and Chapters 5 and 6 in this volume). It is also important to stress that this definition clearly stipulates that a technocrat does not become a “technopol” by assuming a political appointment, as has been argued by others (see, e.g. Williamson 1994). A technopol is an elected politician with policy expertise, who controls important political capital such as political experience and influence within a political organization (Joignant 2011). A separation between expertise and career is crucial for identifying the sources of policy influence. If policy influence is a function of one’s expertise and effort (Alesina and Tabellini 2007), then it is important to be able to identify how much of the outcome is attributable to expertise alone and how much is attributable to effort. Two ministers with identical expertise, such as a PhD in economics, might deliver quite different policies depending on their career objectives—namely, whether they seek re-election, or to go back to their pre-politics career (e.g. in academia or in finance). This chapter provides the first large-n test on the policy effects of technocrat ministers, as opposed to elected partisans, in 13 West European countries since 1980. Utilizing new data on economic and social welfare reforms (Angelova et al. 2018), as well as data on public social spending (OECD 2016), it addresses the following questions: Are technocrat ministers more effective in changing policy than experienced politicians or experts? Are technocrats associated with marketconforming policies? Does it matter which portfolios technocrats are appointed to? And if so, why? The  analysis uses three datasets to test the role of ministers in policy. One dataset codes ministers’ professional and educational expertise, as well as their political and professional career before and after their political appointment (Alexiadou 2016), allowing us to distinguish technocrats from partisans. In addition, this dataset identifies the ministers who are in charge of the portfolios of finance and social welfare. Ministers’ policy effectiveness is tested with a new dataset on policy reforms (Angelova et al. 2018). This dataset counts the number of social welfare and economic reforms per year (Angelova et al. 2018). By utilizing the qualitative indicator of policy measures it is possible to directly test the productivity and effectiveness of ministers, as opposed to the direction of the reform. Finally, the hypothesis that technocrats are appointed to explicitly control government spending is tested utilizing public spending data (OECD 2016).

Technocrats in cabinets and their policy  133 This  chapter is one of the first to test the policy effects of technocrats and expert ministers in European democracies, contributing to a growing literature on the policy effects of individual politicians and of technocrats in particular (see Chapter 6 in this volume; also Alexiadou 2016, Chwieroth 2007, Dargent 2015, Jochimsen and Thomasius 2014, Kaplan 2017). At the same time, it advances the field of executive politics by clearly separating the policy effects of policymakers’ preferences, skills and effort. As such, it further contributes to the debate on the identity of good politicians and effective policymakers (Galasso and Nannicini 2011, Chapter 10 in this volume).

Technocrats or technopols? Who are the most ​effective reformers? Technocrats are appointed to the finance portfolio primarily during major economic and financial crises (Alexiadou and Gunaydin 2019, Kaplan 2017, Schneider 1998). Signalling the government’s pro-market credibility to investors is one of the primary motivations for their appointment (Kaplan 2017, Schneider 1998). Increasing the government’s reform credibility is another. During economic crises, cabinet ministers have to enact policies and adopt reforms that often contradict their electoral promises, their party’s ideology or even their personal convictions. Unlike elected cabinet ministers, who fear the electoral costs of policies they personally introduce, technocrats have not made any electoral promises, they are often appointed to enact their preferred policies for stabilizing the economy, and their professional career is not  subject to electoral approval (Alexiadou and Gunaydin 2019). Indeed, although prime ministers are more likely to appoint both technocrats and technopols as their finance ministers during economic crises, they are twice as likely to appoint technocrats in person-centred than in party-centred electoral systems (Alexiadou and Gunaydin 2019). The  opposite holds for technopols. The  question is whether technocrats are more effective reformers and more likely to push forward unpopular policy reforms than experienced politicians? A policymaker’s effectiveness is a function of her skill and effort (Alesina and Tabellini 2007). Generally, skill is equated with expertise while effort is a function of pay-offs. In the case of cabinet ministers, skill can be broken down to policy expertise and political experience, while effort depends on career objectives (Alexiadou 2016, Alexiadou and Gunaydin 2019). Even though for most scholars expertise is what differentiates technocrats from partisans, it is their non-political career that sets them apart from elected politicians. When appointed to government, elected politicians try to fulfil multiple goals, from advancing their political career to implementing their party’s or their own agenda (Berlinski et al. 2012, Headey 1974, Panebianco 1988). In  contrast, technocrats are appointed with one single mandate: to implement the policy agenda they believe to be the right one for stabilizing or improving the economy. For Alesina and Tabellini (2007), the difference in the objectives of policy tasks is what differentiates bureaucrats from politicians and makes the first more effective policymakers. Similarly, for Blondel (1991),

134  Despina Alexiadou non-elected, non-partisan cabinet ministers should be effective policymakers as they are only marginally concerned about the political implications of their tasks, since they will likely return to their original professions when they leave government. This  is not  to say that partisans do not  want to be effective policymakers or that they do not pursue ideologically motivated policies that carry political costs. Nonetheless, only a minority of partisans can be identified as ideologues (Alexiadou 2016); the average cabinet minister seeks to exhibit capacity in policymaking in light of advancing her political career. Cabinet ministers are cognizant of the fact that unpopular policies can carry heavy personal political costs, even if they are deemed necessary. Balancing the budget during economic crises is certainly one such instance. For former Fine Gael Minister for Social Welfare Gemma Hussey accepting the cuts proposed by the Ministry of Finance would equate to political suicide: “Of course I didn’t go into this job for crucifixion or self-destruction” (Hussey 1990: 207). Being promoted to a cabinet post is not necessarily a stepping stone for further recognition; it can lead to one’s political demise. Whereas for the elected politician being the one to introduce unpopular reforms carries significant personal/political risks, for the technocrat who will go back to her/his prior occupation, the challenge to reform the country’s economic or tax system, for example, carries a low level of personal risk and presents an opportunity to exhibit her/his policy expertise. As noted by former Swedish technocrat Finance Minister Anders Borg, “When I look at other politicians I tend to see myself more as an economist” (Alexiadou and Gunaydin 2019). Cabinet ministers might object to certain policy reforms not only because they fear for their political future, but also because they feel bound by the promises they have made to their constituents. There is increasing experimental evidence showing that elections modify policymakers’ behaviour (Drazen and Ozbay 2015). Technocrats, on the other hand, “start with a blank notebook” (personal interview with former technocrat social welfare minister, June 2011). The fact that technocrats are twice as likely to be appointed in personalistic electoral systems than in party-centred systems during economic crises suggests that technocrats are primarily appointed to address prime ministers’ fear of no reform. Technocrats are also likely to have stronger and more fixed views about the reforms they seek to implement. The tension between policy and office is what differentiates politicians (Headey 1974, Panebianco 1988), and those who have the strongest policy views tend to be the ones with the largest policy influence (Alexiadou 2016). Technocrats are explicitly appointed for their expertise, which often comes with a set of fixed ideas about the nature of the reforms to be adopted. Additionally, technocrats can credibly threaten to not  accept the job, or quit if they are not  allowed to implement their policy agenda, which allows them to wield further influence (Alexiadou 2016). This is exactly what Domenico Siniscalco, the Italian former technocrat finance minister under Silvio Berlusconi, did on 21 September 2005. He resigned only a year after his appointment, because he could not get his 2006 budget approved and because the government protected the central bank governor who was implicated in a scandal (The Economist 2005).

Technocrats in cabinets and their policy  135 Finally, technocrats are experts in their policy domain. Expertise alone can enable cabinet ministers as policymakers to the extent that expertise gives them an advantage over their cabinet colleagues, and even over the prime minister. When technocrats are appointed to the finance portfolio, which is the second most important in a parliamentary government, prime ministers delegate a significant amount of power. Dewan and Hortala-Vallve (2011) show that appointing experts in powerful portfolios entails serious agency risks, even in Westminster systems where prime ministers “fully control” their cabinet. To sum up, technocrats’ expertise, intensity in policy preferences and career incentives should enable them to be effective reformers. This, however, does not mean that technocrat ministers do not face serious hurdles in their reform efforts. First, technocrats have no prior experience in government or in the legislature, and thus lack crucial skills for convincing their colleagues around the cabinet table; communicating effectively with the bureaucrats in their department, and with backbenchers and stakeholder groups; and, last but not least, handling the media (Dowding and Dumont 2009). In addition, technocrats cannot assume the support of the parliamentary party, of public opinion or of important interest groups. To the extent that they are appointed because their preferences diverge from those of the party, they should expect to be attacked on a regular basis. The Greek technocrat social affairs minister, Tasos Giannitsis, was subject to multiple political attacks from his “own” party when trying to reform the pension and social welfare system in Greece. Giannitsis, an economics professor who had been an adviser to Prime Minister Simitis, was appointed with an ambitious policy agenda to make the pension system fiscally sustainable. However, almost none of Giannitsis’s proposals made it to the final bill as they were vetoed by PASOK’s MPs and the trade unions (Lampsias 2001). Party discipline is often a challenge for governing parties, but this is particularly true during economic crises (Herzog and Benoit 2015). In  the midst of the Great Recession, the Greek social-democratic prime minister, George Papandreou, reshuffled his cabinet and appointed Evangelos Venizelos, a party heavyweight and party leadership contender, as his finance minister. According to the Financial Times, “Papandreou yielded to pressure from his fractious Socialist party after threats by deputies to bring the government down over a new austerity package” (Financial Times 2011). While prime ministers might be incentivized to appoint technocrats during periods of economic crisis, it is not clear that technocrats are ultimately able to deliver unpopular reforms. Even if there is willingness by the prime minister to push forward with a set of reform measures, there is a lot of room for slippage. Indeed, a number of recent empirical studies confirm that political experience and party rank increase ministers’ policy influence, in particular when these ministers are in charge of “softer” portfolios such as social welfare (Alexiadou 2016, Jochimsen and Thomasius 2014). Given that technocrats lack the political resources that experienced, elected politicians have at their disposal, their effectiveness will be conditional on the ministerial portfolio to which they are appointed. A  technocrat in the finance

136  Despina Alexiadou portfolio has the de facto power of the second most important job in government, and is in charge of drafting the government’s annual budget (Hallerberg 2004, Hallerberg et al. 2009). In contrast, a technocrat appointed to a softer portfolio (such as that of social welfare) will have a much harder time overcoming resistance from other cabinet colleagues and the backbench. Thus, H1: Technocrat finance ministers should be more effective reformers than experienced partisans. H2: Technocrats in the portfolios of social affairs should be less effective reformers than experienced partisans.

Economic liberalization and technocrats Technocrats and technopols—economists turned politicians—were instrumental in liberalizing the economies of Latin American countries in the 1980s and 1990s by adopting market-conforming policies, also known as the “Washington Consensus” (Kaplan 2014, 2017, Williamson 1993). When it comes to macroeconomic policies, the Washington Consensus maintained that fiscal discipline and tight monetary policy are necessary conditions for competitive and high-growth economies (Williamson 1994). These views, founded on neoclassical economics, reflected the consensus among economists since the latter half of the twentieth century (Christensen 2017, Fourcade 2009, Chapter 2 in this volume). In  Europe, technocratic presence in cabinets was limited during the 1980s.1 However, a significant number of Northern European countries were already enjoying high levels of monetary stability, partly due to influential independent central banks and corporatist bargains between employers and unions (Alexiadou 2012, Crouch 1993, Hall 2013, Mares 2006). Furthermore, the budget-correcting austerity policies of the 1980s and the early ’90s did not seriously threaten the mature European welfare states (Gingrich 2015, Korpi and Palme 2003, Scruggs 2006). Nonetheless, important social welfare and fiscal reforms took place with the aim of liberalizing the labour markets and cutting public pensions to make them fiscally sustainable (Christensen 2017, Gingrich 2015). The policy prescriptions coming out of think tanks, such as the Organization for Economic Cooperation and Development (OECD), attributed slow economic growth to generous welfare payments that created “poverty traps” and to barriers in business restructuring due to employment protection legislation (OECD 1996). Where do the technocrats appointed to the portfolios of finance and economics stand in relation to market-conforming policies? As with partisan politicians, we would expect technocrats’ policy preferences to largely vary with their professional career. After all, even central bankers who have a rather straightforward policy mandate vary in the intensity of their preferences (Adolph 2013, Ainsley 2017). However, it is probably uncontroversial to assume that technocrat finance ministers who are appointed to address a fiscal or monetary crisis are more likely to cut the budgets of high-spending departments and in policy domains that are popular with voters. Health spending and pensions are both policy areas that are

Technocrats in cabinets and their policy  137 particularly hard to reform as they benefit the vast majority of voters across the ideological spectrum (Hausermann 2010). Thus, H3: Technocrat finance ministers should be more effective in cutting public health and pension spending than experienced politicians.

Empirical analysis Independent variables: technocrat ministers of finance and social welfare Finance ministers are the second most powerful actors in parliamentary cabinets, after the prime minister. They often enjoy the prime minister’s support, but not always (Alexiadou 2016). This  is not  the case with most other cabinet ministers, whose policy effectiveness typically relies on the prime minister’s support or the support of the party. As a result, technocrats in the finance portfolio are given significant policy agenda power simply by being appointed to that portfolio. In contrast, technocrats in softer portfolios do not enjoy the same advantage. Their policy effectiveness depends on their ability to convince their colleagues in the cabinet, which is often a function of their party rank and political experience (Alexiadou 2016). Indeed, “spending” ministers have to “fight” for their policy agenda, often against the finance minister (Alexiadou 2016, Alexiadou and Hoepfner 2019, Hallerberg 2004). Second, finance ministers have a specific and relatively narrow set of technical skills, which is mostly concentrated in the field of economics. Identifying the “expert” background for many other portfolios is considerably more difficult. Figure 7.1 summarizes the professional background of finance and social welfare ministers in 18 parliamentary democracies. Interestingly, in both portfolios professors (and teachers) are the largest professional group. However, when we look at technocrats there are important differences between those in the finance and social welfare portfolios. For finance technocrats, defined as those who never held elective office and have policy expertise in economics, finance or academia (Alexiadou 2018, Alexiadou and Gunaydin 2019), more than 40% come from academia and almost 20% worked as economists, which means that about 60% of technocrats are economists. However, for the portfolio of social welfare, which is a lot more diverse and varies in its policy jurisdiction by country and over time, there is an almost perfect split among the top three occupations of ministers: professors, trade union leaders and medical doctors constitute 20% of non-elected social affairs ministers each.2 In  other words, while the coding of technocrats in the finance portfolio is quite straightforward, the coding of technocrats in the social welfare portfolio is the opposite. First, what qualifies as expertise in social welfare? Even though a labour economist is an expert, there are very few of those. What about trade union leaders? Even if they are experts, it is highly unlikely that they are technocrats. Nonetheless, given the small number of technocrats in the social welfare portfolio (see Table 7.1), it is preferable to include former trade union leaders in the operationalization of technocrat social welfare ministers.

Professional Background Prior to Ministerial Appointment (1980–2015) All Finance Ministers

Technocrat Finance Ministers

Professor

Professor

Lawyer Civil Servant Economist

Economist

Local Politics Finance/Banking Executive

Finance/Banking

Full Time Union Living Off Politics Policy Advisor

Lawyer

Blue Collar Journalist Civil Servant

Consultant Suprantional Institution Other

Journalist

Full Time Interest Group Full Time Employers' Organization Military

Policy Advisor

European Parliament Engineer Salaried Employee

Executive

Businessman

0

10

20

30

40

0

10

20

%

30

40

%

Professional Background Prior to Ministerial Appointment (1980–2015) All Social Affairs Ministers

Social Affairs Ministers: Never Elected

Professor Full Time Union

Lawyer Full Time Union Civil Servant Medical Doctor Local Politics Policy Advisor Journalist Living Off Politics Businessman Blue Collar Humanities Unknown Salaried Employee Economist Scientist Executive Military Full Time Interest Group European Parliament Suprantional Institution Engineer Consultant Finance/Banking

0

5

10

15 %

Professor

Medical Doctor

Lawyer

Policy Advisor

Civil Servant

Businessman

Journalist

20

25

0

5

10

15 %

Figure 7.1   Professional background of finance and social affairs ministers

20

25

Technocrats in cabinets and their policy  139 Table 7.1  Frequency of technocrats and experts in the portfolios of finance and social affairs (1980–2012) Country

Finance Finance Finance S. Affairs S. Affairs Social technocrats experts experience technocrats PhDs affairs experience

Austria Belgium Denmark Finland France Germany Greece Italy Netherlands Norway Portugal Spain Sweden Total Average

0.03 0.09 0.00 0.03 0.23 0.00 0.09 0.20 0.00 0.06 0.66 0.00 0.23 0.12

0.77 0.49 0.40 0.34 0.46 0.80 0.60 0.86 0.74 0.43 0.83 0.83 0.69 0.63

0.31 0.69 0.57 0.17 0.09 0.51 0.09 0.26 0.23 0.11 0.00 0.20 0.23 0.27

0.20 0.00 0.03 0.00 0.00 0.00 0.14 0.09 0.17 0.37 0.11 0.00 0.00 0.09

0.29 0.49 0.03 0.57 0.57 0.51 0.26 0.06 0.43 0.14 0.00 0.14 0.00 0.27

0.14 0.34 0.20 0.20 0.17 0.31 0.14 0.14 0.31 0.03 0.06 0.00 0.14 0.17

The ministers’ dataset codes the appointments of finance and social welfare ministers at the start of governments, but it also codes ministerial reshuffles during the life of governments. The data include information on the professional and political backgrounds of finance and employment ministers, collected from multiple official and personal website sources. On average, about 12% of finance ministers are technocrats. Yet, as can be seen in Table 7.1, there is an important variation across Europe, with Portugal having far more technocrats than the other countries (Costa Pinto and Almeida 2017, Costa Pint et al. 2017). The percentage of technocrat social welfare ministers who have never been elected is just 9%. In contrast, 63% of finance ministers are experts, meaning they have a background in academia, economics or banking, and 27% of social welfare ministers have a PhD.3 Finally, just over a quarter of all finance ministers and a sixth of social welfare ministers have cabinet experience, defined as those who served for at least four years in a cabinet post prior to the appointment. To test the policy impact of ministers, the data are transformed to country/year observations. The original dataset identifies one individual minister who is responsible in a policy area. In cases where an election took place or a new cabinet was formed in the middle of a calendar year, it is assumed that the last election or cabinet reshuffle in that calendar year is the one in charge of the policy for that whole calendar year. This coding method takes into account only appointments that lasted more than a few months. The advantage of using country/year data instead of data structured by country/cabinets is that one can clearly identify the year a minister controlled a portfolio and the year a specific policy reform took place.

140  Despina Alexiadou

Dependent variables: economic and social welfare policy reform Technocrats’ policy impact is tested using two sets of variables: one that is qualitative and counts the number of reforms per policy area, and one that measures outcomes, namely public, health and pensions spending as a percentage of GDP. Each set of variables tests different hypotheses. The qualitative indicators capture how effective ministers are as reformers, independently of the direction of the reform, thus testing the first two hypotheses. However, these indicators have the downside that they are only available for 11 out of the 13 countries in the sample, and between 1985 and 2005, thus excluding the years after the 2008 financial crisis. In  contrast, the spending data are available for all 13 countries and for most of the years between 1980 and 2015. In addition, the spending data clearly capture the direction of policy. Therefore, the spending data are utilized to test Hypothesis 3. Two indicators are taken from the policy reform dataset created by Angelova et al. (2018): economic reforms and social welfare reforms. The dataset in total codes 5,600 socioeconomic reforms across 13 Western European countries and over 20  years (1985−2005). After merging the ministerial data with the policy reform data, 11 Western European parliamentary democracies remain in the dataset: Austria, Belgium, Denmark, Finland, France, Germany, Italy, the Netherlands, Portugal, Spain and Sweden, thus losing Greece and Norway.4 These models include a battery of political and economic controls that are known in the literature to predict policy change. The  political controls are: the seat-weighted ideological alternation of the government, measured as the distance between the incumbent government and the average of all governments for the preceding four years (Angelova et al. 2018: 294); the years left in the legislative term; and election years. The expectation is that fewer reforms should take place in an election year while more reforms should happen at the beginning of the legislative term (Hübscher and Sattler 2017), and when the distance between the incumbent government and the status quo is higher. Finally, the left–right ideological position of the prime minister as measured by the Comparative Manifesto Project (RILE) is included. The economic controls are: economic recession, coded as one when GDP is negative, and employment crisis, coded as one when unemployment is higher than 11.6, the sample mean plus one standard deviation (Angelova et al. 2018). Finally, the indicators for Union Density and Wage Bargaining Level (Visser 2013) are included to control for the role of the economic institutions. In terms of the policy outcome variables, three indicators are used: total public spending, spending on health, and spending on public pensions as percentage of GDP. These data are coded by the OECD and were retrieved by the Quality of Government Database (Samani et al. 2010). With these data it is possible to test whether technocrat finance ministers are associated with more cuts in public spending than partisan ministers. Empirical models Two different estimation methods are employed for the two sets of dependent variables: a negative binomial model for the qualitative indicators of policy

Technocrats in cabinets and their policy  141 reform, and an error correction model for the spending indicators. Negative binomial estimation is used for over-dispersed count data when the conditional variance exceeds the conditional mean. Policy reforms are count variables. For each country/year, the total number of reforms per policy are counted. The negative binomial regression has the same structure as the Poisson regression, but with an extra parameter to model the overdispersion. To control for country heterogeneity the errors are clustered by country, and country fixed effects are included. Error correction models (ECM) directly model the dynamic effects in the data—namely, how long interventions have an effect on the dependent variable. In ECM models the dependent variable is in changes, while all regressors are in both changes and lagged levels. This  model is appropriate for spending data, which are stationary and balanced (Keele et al. 2016). Country fixed effects are included. Empirical results Tables  7.2 and 7.3 test Hypothesis 1 and 2, respectively. The  results provide weak support to both hypotheses. According to Table 7.2, technocrat finance Table 7.2  The role of technocrat finance ministers in economic reforms

Recession Job Crisis Alternation Years To Election Election Year Union Density LEVEL Technocrat Expert

(1)

(2)

(3)

Economic reforms

Economic reforms

Economic reforms

1.25 (0.24) 2.48*** (0.26) 1.24** (0.12) 1.02 (0.05) 0.74*** (0.06) 0.99 (0.01) 1.02 (0.10) 1.52* (0.35)

1.27 (0.24) 2.47*** (0.25) 1.21* (0.13) 1.01 (0.06) 0.75*** (0.07) 1.00 (0.01) 1.04 (0.10)

1.29 (0.25) 2.42*** (0.25) 1.22* (0.13) 1.01 (0.06) 0.75*** (0.06) 1.00 (0.01) 1.04 (0.10)

Experienced Constant Observations

5.25*** (3.22) 240

Robust standard errors in parentheses ***p