Policy Analysis in France

INTERNATIONAL LIBRARY OF POLICY ANALYSIS
SERIES EDITORS:
IRIS GEVA-MAY & MICHAEL HOWLETT
POLICY ANALYSIS IN
France
Edited by Charlotte Halpern, Patrick Hassenteufel

and Philippe Zittoun
POLICY ANALYSIS
IN FRANCE
International Library of
Policy Analysis
Series editors: Iris Geva-May and Michael Howlett,
Simon Fraser University, Canada
This major new series brings together for the first time a detailed
examination of the theory and practice of policy analysis systems
at different levels of government and by non-governmental actors
in a specific country. It therefore provides a key addition to
research and teaching in comparative policy analysis and policy
studies more generally.
Each volume includes a history of the country’s policy analysis which

offers a broad comparative overview with other countries as well as the
country in question. In doing so, the books in the series provide the
data and empirical case studies essential for instruction and for further
research in the area. They also include expert analysis of different
approaches to policy analysis and an assessment of their evolution and
operation.
Early volumes in the series will cover the following countries:

Australia • Brazil • China • Czech Republic • France • Germany •
India • Israel • Netherlands • New Zealand • Norway •
Russia • South Africa • Taiwan • UK • USA
and will build into an essential library of key reference works. The series
will be of interest to academics and students in public policy, public
administration and management, comparative politics and government,
public organisations and individual policy areas.
It will also interest people working in the countries in question
and internationally.
In association with the ICPA-Forum and Journal of Comparative Policy Analysis.

See more at http://goo.gl/raJUX
POLICY ANALYSIS
IN FRANCE
Edited by Charlotte Halpern, Patrick Hassenteufel and
Philippe Zittoun
International Library of Policy Analysis, Vol 11

First published in Great Britain in 2018 by
Policy Press North America office:
University of Bristol Policy Press
1-9 Old Park Hill c/o The University of Chicago Press
Bristol BS2 8BB 1427 East 60th Street
UK Chicago, IL 60637, USA
+44 (0)117 954 5940 t: +1 773 702 7700
pp-info@bristol.ac.uk f: +1 773 702 9756
www.policypress.co.uk sales@press.uchicago.edu
www.press.uchicago.edu
© Policy Press 2018
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library.
Library of Congress Cataloging-in-Publication Data
A catalog record for this book has been requested.
ISBN 978-1-4473-2421-8 hardcover
ISBN 978-1-4473-4739-2 ePub
ISBN 978-1-4473-4740-8 Mobi
ISBN 978-1-4473-2423-2 ePdf
The right of Charlotte Halpern, Patrick Hassenteufel and Philippe Zittoun to be identified as
editors of this work has been asserted by them in accordance with the Copyright, Designs and
Patents Act 1988.
All rights reserved: no part of this publication may be reproduced, stored in a retrieval system,
or transmitted in any form or by any means, electronic, mechanical, photocopying, recording,
or otherwise without the prior permission of Policy Press.
The statements and opinions contained within this publication are solely those of the editors
and contributors and not of the University of Bristol or Policy Press. The University of Bristol
and Policy Press disclaim responsibility for any injury to persons or property resulting from any
material published in this publication.
Policy Press works to counter discrimination on grounds of gender, race,
disability, age and sexuality.
Cover design by Qube Design Associates, Bristol
Front cover: image kindly supplied by istock
Printed and bound in Great Britain by CPI Group (UK) Ltd,
Croydon, CR0 4YY
Policy Press uses environmentally responsible print partners
Contents
Notes on contributors vii

Editors’ introduction to the series xiii
one Policy analysis in France: introduction 1
Charlotte Halpern, Patrick Hassenteufel and Philippe Zittoun
Part One: The styles and methods of public policy analysis
two On the path to public policy analysis: 31

an ‘administrative science’ between reform and academy
Renaud Payre and Gilles Pollet
three The emergence of modern policy analysis in France 47
Fabrice Hamelin
four Recent developments within French policy studies 63
Jacques de Maillard and Andy Smith
five Methods of French policy studies 79
Claire Dupuy and Philippe Zittoun
Part Two: Policy analysis by governments
six Civil servants and policy analysis in central government 101

Emilie Biland and Natacha Gally
seven Policy analysis in French local government 119
Alain Faure and Emmanuel Négrier
eight Beyond weakness: policy analysis in the French parliament 137
Olivier Rozenberg and Yves Surel
Part Three: Committees, public inquiries, and consultants
nine Public inquiries, committees 157

Cécile Blatrix and Guillaume Gourgues
ten Management consultants as policy actors 175
Julie Gervais and Frédéric Pierru
Part Four: Parties, interest groups, research institutes and think tanks
eleven The field of state expertise 191

Mathieu Hauchecorne and Etienne Penissat
twelve Political parties and think tanks: policy analysis oriented toward 209
office-seeking
Camilo Argibay, Rafaël Cos and Anne-Cécile Douillet
v
Policy analysis in France
thirteen Economic interest groups and policy analysis in France 225

Guillaume Courty and Marc Milet
fourteen NGOs, civil society and policy analysis: from mutual disinterest 243
to reciprocal investment
Laurie Boussaguet and Charlotte Halpern
fifteen Trade union expertise in public policy 261
Sophie Béroud and Jean-Marie Pernot
Part Five: Academic policy analysis
sixteen Economics and policy analysis: ‘from state to market’? 279

Daniel Benamouzig and Frédéric Lebaron
seventeen The academic world of French policy studies: training, 295
teaching and researching
Patrick Hassenteufel and Patrick Le Galès
eighteen Public policy analysis in France: from public action to political power 313
Patrice Duran
Index 333
vi
Notes on contributors
Camilo Argibay holds a PhD in political science from the University of Lyon.
His doctoral research focused on students’ activism and he is now working on
the role of think tanks in the French political system.
Daniel Benamouzig is a research professor in sociology at the National Centre

for Scientific Research (CNRS) and Sciences Po, Centre de Sociologie des
Organisations (CSO). He is also Director of the Institute of Public Health of
the National Alliance for Research in Biomedicine and Health (AVIESAN) and
Chair of the Committee for Social Science in the Higher Authority for Health
(Haute Autorité de Santé). His has authored many publications about public
health, economic policy and institutional governance, and his research contributes
to economic sociology and political sociology of healthcare.
Sophie Béroud is an associate professor in political science at the University of

Lyon 2 and member of Triangle Centre for research (UMR 5206, ENS-LSH,
IEP, Lyon 2). Her work focuses on the current transformations of French and
Spanish trade unionism. Among her latest publications: ‘Economic crisis and
social protest in Spain: Labor unions and social movements at odds’, Critique
internationale, 65, 2014, pp. 27–42; ‘The organization of the unemployed in
Spain: Local and fragmented dynamics’ in Didier Chabanet and Jean Faniel (eds)
The mobilization of the unemployed in Europe: From acquiescence to protest? (Palgrave
Macmillan, 2012, pp. 155–74).
Emilie Biland is an associate professor of political science at the University of

Rennes 2 (France) and a research fellow at ARENES (UMR 6051). Her research
focuses on the part played by law and courts in public policy in a comparative
perspective.
Cécile Blatrix is a professor of political science at AgroParisTech, Paris. Her

works focuses on public participation.
Laurie Boussaguet is full professor of political science at the University of

Rouen and visiting fellow at the Robert Schuman Centre for Advanced Studies
(European University Institute, Florence, Italy). Her research focuses on policy
analysis, symbolic policies, transnational convergence, European governance and
participatory democracy in the EU. She is the author and editor of many books
and peer-reviewed articles about public policies and policy studies. Her latest
publication, ‘The politics of symbols’, was published in Parliamentary Affairs (2017).
Rafaël Cos is a PhD student in political science at Lille University, member of

the Lille Centre for European Research on Administration, Politics and Society
vii
(CERAPS). His research deals with the programmatic work within the French
Socialist Party.
Guillaume Courty is a professor of political science at Lille University, Centre

for European Research on Administration, Politics and Society (CERAPS). His
research focuses on the lobbyist’s profession and the transport sector.
Anne-Cécile Douillet is professor of political science at Lille University, member

of the Lille Centre for European Research on Administration, Politics and Society
(CERAPS). She is a specialist in policy process analysis and local government,
her works dealing mainly with local development policies and public security
policies. She recently published, with Rémi Lefebvre, Sociologie politique du pouvoir
local (Armand Colin, 2017).
Claire Dupuy is associate professor of political science at Sciences Po Grenoble

– Pacte (France). She specialises in comparative public policy with a focus on
state transformations and regionalisation processes in western Europe. She is also
interested in policy feedbacks and the ways in which policy changes have an impact
on citizens’ (dis)affection toward politics. Empirically, she conducts research on
education and social policy in France, Germany and Belgium.
Patrice Duran is a professor at École Normale Supérieur (ENS) de Paris Saclay

and member of the Institut des Sciences sociales du Politique (ISP), ENS Paris
Saclay/CNRS. He is also President of the Conseil National de l’Information
Statistique (CNIS). His main fields of research are political sociology, sociological
theory, state analysis, comparative analysis, territorial governance, sociology of
law and evaluation of public policies.
Alain Faure is National Centre for Scientific Research (CNRS) senior researcher
at PACTE, University of Grenoble Alpes. His last fields of research concern
the socialisation of local elected representatives in a comparative perspective
(France, Canada, Italy, Japan). He published in 2016 Des élus sur le divan (Presses
universitaires de Grenoble) and co-edited in 2017 with E. Négrier La politique à
l’épreuve des émotions (Presses universitaires de Rennes).
Natacha Gally is an associate professor of political science at Université Paris II-

Panthéon Assas and a research fellow at the Center for the Studies and Research
of Administrative and Political Sciences (CERSA–CNRS) (UMR 7106). Her
research deals with the transformations of the state with a specific interest in
senior civil service reforms and administrative labour markets in a comparative
perspective.
Julie Gervais is an associate professor in political sociology at Sorbonne University.

She works on the political sociology of the state, elites and public policy-making.
viii
She addresses topics including French high civil servants (training, careers,
revolving-door phenomena), public sector consultants and the managerialisation
of the state, lobbyists on campaign trails and the question of influence. Her
current research project is on corporate political influence orchestrated by the
direct mobilisation of popular participation.
Guillaume Gourgues is associate professor in political science (University of

Bourgogne Franche-Comté, France). His work focuses on public participation
and industrial democracy through a multilevel sociological approach. He notably
published ‘Studying regions as spaces for democracy: A political sociological
approach’ (Regional and Federal Studies, 20(3), 2010) and Les politiques de démocratie
participative (Presses universitaires de Grenoble, 2013).
Charlotte Halpern is an associate research professor in political science at Sciences

Po, Centre d’études européennes et de politique comparée (CEE) CNRS, in
Paris, France. She specialises in comparative policy studies, urban studies and EU
governance. Among her latest publications, she has recently co-edited two special
issues: (with P. Graziano) ‘EU governance in times of crisis: Inclusiveness and
effectiveness beyond the ‘hard’ and ‘soft’ law divide’, Comparative European Politics,
14(1), 2016; (with C. Belot and L. Boussaguet), ‘Gouverner (avec) l’opinion
publique européenne’, Politique européenne, 54(4), 2016.
Fabrice Hamelin is associate professor in political science at Université Paris-

Est Créteil, LIPHA (EA 7373) where he teaches policy analysis and governance
studies. In his research, he focuses on the comparative analysis of transport security
and safety policy-making, with some recent work on tools of government and
policy design in mobility and transport, as well as the history of applied transport
research in Europe.
Patrick Hassenteufel is a professor of political science at the University of

Versailles and Sciences Po Saint-Germain-en-Laye, member of the Printemps
(CNRS research centre). He has authored many publications about comparative
health policy, the transformations of European welfare states and sociological
actor-centred policy analysis (among them a handbook on policy analysis: Sociologie
politique, l’action publique, Armand Colin, 2008 [2nd edn, 2011]). He is the co-
editor in chief of Gouvernement et action publique.
Mathieu Hauchecorne is an assistant professor in political science at Paris

8 University and a member of the CRESPPA-LabToP. His research interests
centre on the sociology of intellectuals and expertise, intellectual history, and
the analysis of policy transfers and the circulation of ideas. His research on the
French reception of Rawls and US theories of justice has led to several articles
in academic journals and a book is forthcoming with CNRS éditions.
ix
Frédéric Lebaron is professor at École normale supérieure Paris-Saclay, where

he heads the department of social sciences. He also teaches at Sciences Po-
Saint-Germain-en-Laye. He is a member of two research groups: Institutions
Dynamiques Historiques des Economies et des Sociétés and Laboratoire
Printemps. He has co-edited (with B. Le Roux) La méthodologie de Pierre Bourdieu
en action (Dunod, 2015), and, with M. Grenfell, Bourdieu and Data Analysis (Peter
Lang, 2014).
Patrick Le Galès is CNRS research professor in politics and sociology at Sciences

Po, Centre d’études européennes et de politique comparée (CEE), CNRS. He
is also the Dean of Sciences Po Urban School. His recent publications include
Reconfiguring European states in crisis, with D. King (Oxford University Press,
2017), Globalising minds, roots in the city, European urban middle classes (with A.
Andreotti, J. Moreno Fuentes) (Wiley, 2015), What is governed in Mexico City?
(forthcoming, 2018).
Jacques de Maillard is professor of political science at the University of Versailles-

Saint-Quentin, deputy-director of the Cesdip. His interests lie in the questions
of local governance of security, the comparative study of policing in Western
countries and theories of public policy analysis. He has recently published (with S.
Savage), ‘Policing as a performing art? The contradictory nature of contemporary
police performance management’, Criminology and Criminal Justice, 2017, pp. 1-13
(online version), and with S. Roché, ‘Studying policing comparatively: obstacles,
preliminary results and promises’, Policing & Society, 2016, pp. 1-13.
Marc Milet is assistant professor in political science at University Paris Panthéon

Assas, and a member of the Center for the Studies and Research of Administrative
and Political Sciences (CERSA – CNRS). His research focuses on legislative
studies and on small business interest groups in France and within the European
Union, which is the topic of his forthcoming book: Théorie critique du lobbying:
L’UEAPME et la revendication européenne des petites et moyennes entreprises (Paris:
Logiques politiques, L’Harmattan).
Emmanuel Négrier is CNRS senior researcher at the South European Centre

for Political Studies (CEPEL), University of Montpellier. His main fields of
research are territorial reforms, comparative politics and policies, cultural policy
and political behaviour. He recently published, together with Alain Faure, La
politique à l’épreuve des émotions (Presses universitaires de Rennes, 2017).
Renaud Payre is professor of political science at the Université de Lyon, Sciences

Po Lyon, Laboratoire Triangle and director of Sciences Po Lyon since 2016. He is
a specialist of the socio-historical approach in public policy analysis. He recently
published, with Alistair Cole, Cities as political object: Historical evolution, analytical
categorisations and institutional challenges of metropolitanisation (Edward Elgar, 2016).
x
Etienne Penissat is research fellow at the National Centre for Scientific Research
(CNRS) and member of the Centre for European Research on Administration,
Politics and Society (CERAPS), Lille University. His research interests focus on
the expertise, production and use of statistics in public policy. He has published
on the transformations of these intellectual productions at the Ministry of Labour
since 1945.
Jean-Marie Pernot is senior researcher and political scientist at Institut de

Recherches Economiques et Sociales (IRES) and Centre d’Histoire Sociale
du XX° siècle. He also teaches at the university of Lyon 2 and at the CELSA
(Paris 4- Sorbonne). His main research domains are the trade unions and social
movements in historical and international perspectives. Among his most recent
publications is ‘France’s trade unions in the aftermath of the crisis’, in S. Lehndorff,
H. Dribbusch and T. Schulten (eds) Rough waters: European trade unions in a time
of crises (European Trade Union Institute, 2017).
Frédéric Pierru holds a PhD in political science and is sociologist and a

CNRS Research Fellow, affiliated with the Centre for European Research on
Administration, Politics and Society (CERAPS) (UMR 8026). At the crossroads
of sociology of the state, public action and the medical field, his works and articles,
written alone and in collaboration with others, deal with the processes of reform
of healthcare systems, the restructurings of the French health administration
(health protection and epidemiological supervision) as well as the transformations
of hospital medicine and medical mobilisations under the effect of managerial
actors, logics and instruments.
Gilles Pollet is a professor of political science at the University of Lyon/Sciences

Po Lyon/UMR Triangle and Director of the Institut des Sciences de l’Homme
(ISH). He is working on public policy analysis in socio-historical perspectives,
especially on the development and transformations of welfare states and local
governments, and also on ‘sciences de gouvernement’. His publications include
Socio-histoire de l’action publique (La Découverte, 2013), with Renaud Payre.
Olivier Rozenberg is an associate professor of political science in Sciences Po, at

the Centre d’études européennes et de politique comparée (CEE), CNRS, Paris,
France. He specialises in EU politics and legislative studies. He has co-edited
volumes on parliamentary roles (Routledge, 2012), questions (Routledge, 2012)
and on national parliaments in the EU (Palgrave, 2015).
Andy Smith is research professor at the Centre Emile Durkheim, University

of Bordeaux. A specialist of political economy, his latest book is The politics of
economic activity (Oxford University Press, 2016). He is currently studying the
arms industry in France and the UK.
xi
Yves Surel is professor of political science at the University Paris 2 (Panthéon-

Assas). His research focuses on policy analysis and comparative politics. He
recently published La science politique et ses méthodes (Armand Colin, 2015). He
has coedited, with Jacques de Maillard, Les politiques publiques sous Sarkozy (Presses
de Sciences Po, 2012), and with Yves Mény, Democracies and the populist challenge
(Palgrave, 2002).
Philippe Zittoun is research professor of political science at the LAET-

ENTPE University of Lyon, General Secretary of the International Public
Policy Association and Chair of the Research Committee ‘Public Policy and
Administration’ of IPSA. His research is about the policy process, its political
dimension and the role of meaning inside. His most recent books are: Contemporary
Approaches to Public Policy, edited with B. Guy Peters (Palgrave, 2016), and The
Political Process of Policymaking (Palgrave, 2014). He is on the editorial board of
the Policy Studies Journal, Critical Policy Studies, Journal of Policy Research, Journal
of Comparative Policy Analysis and Policy and Society and he is coordinator of the
International Conference on Public Policy.
xii
Editors’ introduction to the series
Professor Iris Geva-May and Professor Michael Howlett, ILPA series editors
Policy analysis is a relatively new area of social scientific inquiry, owing its origins
to developments in the US in the early 1960s. Its main rationale is systematic,
evidence-based, transparent, efficient, and implementable policy-making. This
component of policy-making is deemed key in democratic structures allowing
for accountable public policies. From the US, policy analysis has spread to other
countries, notably in Europe in the 1980s and 1990s and in Asia in the 1990s and
2000s. It has taken, respectively one to two more decades for programmes of public
policy to be established in these regions preparing cadres for policy analysis as a
profession. However, this movement has been accompanied by variations in the
kinds of analysis undertaken as US-inspired analytical and evaluative techniques
have been adapted to local traditions and circumstances, and new techniques
shaped in these settings.
In the late 1990s this led to the development of the field of comparative policy
analysis, pioneered by Iris Geva-May, who initiated and founded the Journal
of Comparative Policy Analysis, and whose mission has been advanced with
the support of editorial board members such as Laurence E. Lynn Jr., first co-
editor, Peter deLeon, Duncan McRae, David Weimer, Beryl Radin, Frans van
Nispen, Yukio Adachi, Claudia Scott, Allan Maslove and others in the US and
elsewhere. While current studies have underlined differences and similarities in
national approaches to policy analysis, the different national regimes which have
developed over the past two to three decades have not been thoroughly explored
and systematically evaluated in their entirety, examining both sub-national and
non-executive governmental organisations as well as the non-governmental sector;
nor have these prior studies allowed for either a longitudinal or a latitudinal
comparison of similar policy analysis perceptions, applications, and themes across
countries and time periods.
The International Library for Policy Analysis (ILPA) series fills this gap in the
literature and empirics of the subject. It features edited volumes created by experts
in each country, which inventory and analyse their respective policy analysis
systems. To a certain extent the series replicates the template of Policy Analysis
in Canada edited by Dobuzinskis, Howlett and Laycock (Toronto: University of
Toronto Press, 2007).
Each ILPA volume surveys the state of the art of policy analysis in governmental
and non-governmental organisations in each country using the common template
derived from the Canadian collection in order to provide for each volume in the
series comparability in terms of coverage and approach.
Each volume addresses questions such as: What do policy analysts do? What
techniques and approaches do they use? What is their influence on policy-making
in that country? Is there a policy analysis deficit? What norms and values guide
xiii
the work done by policy analysts working in different institutional settings?

Contributors focus on the sociology of policy analysis, demonstrating how analysts
working in different organisations tend to have different interests and to utilise
different techniques. The central theme of each volume includes historical works
on the origins of policy analysis in the jurisdiction concerned, and then proceeds
to investigate the nature and types, and quality, of policy analysis conducted by
governments (including different levels and orders of government). It then moves
on to examine the nature and kinds of policy analytical work and practices found
in non-governmental actors such as think tanks, interest groups, business, labour,
media, political parties, non-profits and others.
Each volume in the series aims to compare and analyse the significance of the
different styles and approaches found in each country and organisation studied,
and to understand the impact these differences have on the policy process.
Together, the volumes included in the ILPA series serve to provide the basic
data and empirical case studies required for an international dialogue in the area
of policy analysis, and an eye-opener on the nuances of policy analysis applications
and implications in national and international jurisdictions. Each volume in the
series is leading edge and has the promise to dominate its field and the textbook
market for policy analysis in the country concerned, as well as being of broad
comparative interest to markets in other countries.
The ILPA is published in association with the International Comparative Policy
Analysis Forum, and the Journal of Comparative Policy Analysis, whose mission is
to advance international comparative policy analytic studies. The editors of each
volume are leading members of this network and are the best-known scholars
in each respective country, as are the authors contributing to each volume in
their particular domain. The book series as a whole provides learning insights for
instruction and for further research in the area and constitutes a major addition
to research and pedagogy in the field of comparative policy analysis and policy
studies in general.
We welcome to the ILPA series Volume 11, Policy Analysis in France, edited
by Charlotte Halpern, Patrick Hassenteufel and Philippe Zittoun, and thank
the editors and the authors for their outstanding contribution to this important
encyclopedic database.
Iris Geva-May
Professor of Policy Studies, Baruch College at the City University of
New York, Professor Emerita Simon Fraser University; Founding President
and Editor-in-chief, International Comparative PolicyForum and Journal of
Comparative Policy Analysis
Michael Howlett
Burnaby Mountain Professor, Department of Political Science, Simon Fraser
University, and Yong Pung How Chair Professor, Lee Kuan Yew School of
Public Policy, National University of Singapore
xiv
ONE
Policy analysis in France: introduction

Charlotte Halpern, Patrick Hassenteufel and Philippe Zittoun
Introduction
This book lays the foundation for a more systematic understanding of policy
analysis in France.1 In the French context, understanding ways in which
knowledge of and for policies is produced within and outside the state raises
two issues that we collectively address in this volume: explaining the process
by which studies for policy process have been strictly separated from the study
of policy process and providing some explanation as to why this fundamental
distinction still holds – even though it was regularly challenged by successive
generations of scholars and practitioners. We argue that in the French context,
this remaining divide results from the specificity of the politico-bureaucratic
system, the structuring of the academic sphere as well as the functioning of the
policy-making process.
One of the main difficulties was to define what kind of knowledge can be
considered as ‘policy analysis’ and what cannot (Hassenteufel and Zittoun, 2017).
In the French context, civil servants were the first to develop policy analysis as
an autonomous field of expertise within the state apparatus. From the eighteenth
century onwards, several ‘Grand Corps’ contributed to mapping out the extent of
the state’s intervention in various policy areas as well as providing some suggestions
on how it could be enhanced. One of the most internationally-known examples
is the comparative study done by Tocqueville on prison policies in France and
the United States in 1833 (De Beaumont and De Tocqueville, 1845). Since this
period and until the 1980s, policy analysis was mainly developed as practitioner’s
know-how within the state, that is, as studies for policies, and only drew on academic
expertise on rare occasions.
It was only much later on, during the 1970s and 1980s, that policy studies
emerged as an academic field within political science and administration studies
(Leca and Muller, 2008). They favoured the development of a comprehensive
knowledge that would enable them to grasp the policy process and the role of
bureaucratic elites in producing policy studies for the policy process (Zittoun
and Demongeot, 2010). This historically-situated process explains why and
how ‘policy studies’ is now understood in France – and in most public policy
textbooks – as an academic field and not as applied research.
Together with other books published in the series on ‘policy analysis’, this
volume brings together a detailed examination of the theory and practice of
1
policy analysis systems in the French context. It considers policy studies as a field
of both academic research – across various disciplines (political science, sociology,
public law, management and economics) – and practitioners’ knowledge – in
a large number of organisations in the politico-administrative system, across
society and across levels of governance (parliament, bureaucracies, trade unions,
political parties, think tanks etc). In doing so, it raises the following questions:
how is it produced? By whom is it produced and in what format? Is it considered
as a strategic policy resource, and if so, what uses of policy analysis have
developed over time? At what stages of the policy process is it primarily used?
Is it preferably in order to provide policymakers with knowledge and evidence,
during implementation or as a way to evaluate existing policies? To what extent
do fields of academic research and practitioners’ knowledge overlap or compete
with one another? These questions confirm the need to consider these forms of
expertise for and about policy-making in their diversity.
This book critically addresses the process by which studies for policy process
have been strictly separated from the study of policy process in the French context.
In order to do so, it brings together scholars who are widely recognised as major
contributors to this academic field both in France and beyond. Similar to other
volumes in this series, this book is divided into five sections and 18 chapters,
which all address the on-going tension between the study of and for policy in
various organisations (think tanks, political parties, interest groups etc) and arenas
(parliament, local government etc).2 Yet, they also show that continued exchanges
and relations have ensured the mutual enrichment between both spheres in the
production of knowledge on the policy process. Through the focus on policy
studies, this volume demonstrates why and how the fundamental distinction
between knowledge for and of policy process still holds in the French context,
even though it was regularly challenged by successive generations of scholars
and practitioners. It contributes to comparative studies about policy analysis,
by providing an understanding of what it means, how it is used and how it is
studied in the French context. It also contributes to international debates among
academics and practitioners about policy analysis.
This introductory chapter begins with a brief overview of what policy studies
means and how it is studied in the French context. Then the bulk of the
introduction highlights and provides some explanation for the enduring gap
between academic knowledge and policy practices, which characterises policy
analysis in the French context.3 Last but not least, we discuss the added value of
policy studies for understanding state restructuring and policy developments in
France. In the remaining and fourth section, we introduce the book outline in
mor detail.
2
Introduction
Policy analysis: what it means and how it is studied in the French

context
The notions of policy analysis (in French ‘analyse des politiques publiques’) in
France are ambiguous, which causes difficulties when attempting to define what
policy analysis is and what it is not. This ambiguity is related to the discipline’s
early developments. In the French context, the invention of policy studies in
the academic sphere resulted from the transfer of a series of concepts, ideas and
models that were originally developed in the United States in order to study the
policy process. It was, however, called ‘policy analysis’.
Policy analysis: what it means in the French context
As a concept, policy analysis has already been the topic of heated scholarly debates
in the United States in reference to knowledge about problem-solving – or the
problem-solving approach – and those practices aiming at producing knowledge
for policy-making (Bardach, 2012). Policy analysis should therefore not be
confused with the notion of policy process studies, which holds a more distanced
and comprehensive perspective on the policy process (Zittoun and Peters, 2016).
Nevertheless, the definition of policy analysis remains vaguely defined according
to Wildavsky, and he repeatedly challenged the epistemological status of policy
analysis as a science and an academic field, as opposed to a practical and applied
field of knowledge. In this understanding, policy analysis is not considered a
science but, rather, an art and a craft because ‘without art, analysis is doomed
to repetition; without craft, analysis is unpersuasive’ (Wildavsky, 1987, 389),
thus making it impossible for policy analysis to be monopolised by academics
(Majone, 1992).
Even though French scholars used the term ‘policy analysis’ when referring to a
new academic field in the 1970s, they were primarily inspired by scholarly work
in the United States which was undoubtedly rooted in policy process studies.
There, it was developed from the 1970s onwards as a field of academic research
in political science in which the stated aim was to understand and explain the
functioning of the policy process (Jones, 1970; Lindblom and Woodhouse, 1992;
Anderson, 1979). Interestingly, the authors to which French scholars mostly
referred to in order to develop their own understanding of ‘Analyse des Politiques
Publiques’ are Jones and Lindblom. Yet the former argues against using the concept
of policy analysis and against creating too great a distance from it when studying
the policy process, whereas the latter’s well-known incrementalist model was
initially misunderstood by French scholars, since it was intended precisely as an
analytical and a normative model for solving policy problems.
Nevertheless, the very use of the notion of ‘policy’ as the uncontested way
to designate the object of policy studies – as has been the case in the United
States since 1945 – is also controversial in France. The notion of ‘policy’ was first
translated by Raymond Aron as ‘la politique’4 (Lasswell et al, 1951), which in
3
French refers without distinction to policy, politics and polity (Lasswell, 1951).
Although Lasswell argued that ‘the word “policy” is commonly used to designate
the most important choices…[It] is free of many of the undesirable connotations
clustered about the word political which is often believed to imply “partisanship”
or “corruption”’ (Lasswell, 1951, 2), this was lost in translation. A few decades
later, when policy studies emerged as an academic discipline in France during the
1980s, ‘policy’ was translated and discussed as ‘politiques publiques’ (that is, ‘public
policy’),5 due to its importers’ interest in the understanding of the functioning
of the state and its interventions. Even though the term would appear to be a
literal translation from English, the French appellation implies that policies are
confined to the public domain.
In this volume, and in order to avoid any confusion, we will use, on the one
hand, ‘policy studies’ in order to refer to the knowledge about public policy, which
includes ‘policy analysis’ as a knowledge for policy and is mainly produced, in
the French context, by practitioners. On the other hand, ‘policy process studies’
will be used to refer to the knowledge on the policy process which is mainly
produced in the academic sphere.
The invention of policy process studies in the French context
The emergence of policy studies is closely related to shortcomings of administrative

sciences some 100 years after its creation, in terms of both its institutionalisation
as an academic discipline and as an autonomous field of policy knowledge. In
Chapter Two, Renaud Payre and Gilles Pollet highlight the dominant role of
law studies and practitioners, in contrast to the study of administration, which
was considered an ‘art’, that is, a body of know-how and techniques used in the
training of civil servants.
By contrast, during the 1970s, policy studies strategically developed outside
universities, in research and training institutions such as National Centre for
Scientific Research (Le Centre national de la recherche scientifique [CNRS]) research
centres and the Instituts d’Etudes Politiques in Paris and Grenoble. This was
considered a way to gain some autonomy vis-à-vis Administrative Law. Drawing
on transnational – mainly transatlantic – exchanges for some of them, and coming
from different academic backgrounds, for example, sociology (Michel Crozier,
Jean-Claude Thoenig, Jean-Gustave Padioleau) and political science (Yves Mény,
Yves Barel, Jean Leca, Lucien Nizard), these scholars shared a common interest
in policy studies as an academic field of expertise dedicated to the study of the
state apparatus and its policies (Jobert and Muller, 1987; Dupuy and Thoenig,
1983; Crozier and Friedberg, 1977; Crozier and Thoenig, 1975; Padioleau, 1982).
As any process of diffusion and transfer, this import process was not agentless
and led to the careful selection of some concepts and methods. Through the
import of concepts such as bureaucratic dysfunction, bounded rationality, trial
and error and incrementalism, they critically addressed the role of the state and
sought to increase their understanding of the policy process. Concomitantly
4
Introduction
to this selective import process, they sought to create new concepts in order
to develop an approach mainly influenced by a sociological perspective on the
state, the bureaucracy and the policy process, as well as to explore the governing
of French society. In doing so, they distinguished themselves from, on the one
hand, practitioners and policy practice, and on the other hand, from a Marxist
tradition of analysing the role of the state (Leca and Muller, 2008).
In these early days, ‘politique publique’ was understood as an abstract concept,
whereas identifying the policy process’ main features was best achieved through a
thorough empirical analysis as opposed to theory building. This was considered a
prerequisite to disentangling the bundle of multiple policy decisions, budgets and
regulations. This fuzzy and elusive enough notion – developed by Jean-Claude
Thoenig – nevertheless successfully gathered under the same label a large variety
of existing approaches and methods (Grawitz et al, 1985; Jobert and Muller, 1987)
and ‘l’analyse des politiques publiques’ was acknowledged as a common banner for
everything related to policy studies, including policy analysis, policy process
studies, policy design, public problem and policy outcome studies. It should be
noted that, in these early days, these scholars were particularly anxious to explore
the policy process through its various stages, with a clear focus on the role of
actors and very few references to scholarly work produced in the US.6 In early
textbooks (Mény and Thœnig, 1989; Muller and Surel, 2000), policy analysis
refers exclusively to an academic subfield of political science on policy process.
As such, it unequivocally rejects any problem-solving orientation and seeks to
understand and explain the policy process. A few years later, Bruno Jobert and
Pierre Muller completed this definition in their seminal analysis of The state
in action (1987) by highlighting the importance of representations and society
visions to the definition of public policy. They argued that governments shape
and transform existing social, political and economic structures as well as specific
policy fields by strategically using norms and ideas. From then on, what was later
addressed as the ‘French school of public policy process studies’ investigated ‘the
social image, the representation of the system on which they [governmental actors]
want to act’ (Muller, 2015), alongside policy decisions, budget and regulations.
This understanding remained dominant in academic debates until the 1990s.
The institutionalisation of policy studies as part of political science
The notion of ‘public policy’ and the French concept of ‘policy analysis’, as
defined during these early years in the academic sphere, were challenged by
successive generations of scholars through the import of concepts and methods
from other countries or research traditions. The largest share of critics argued that
the term ‘politiques publiques’ induced strong bias during empirical observation by
pre-supposing the existence of a coherent sets of norms, tools and beliefs, by an
overly state-centred approach that underestimated other levels of government as
well as non-state actors, and by its blindness to the political dimensions of policy-
making (that is the politics) (Gaudin, 2007; Leca et al, 2003). A new generation
5
of scholarly work highlighted the limits of approaches that overestimated the

robustness of national policy models (Jobert, 1994; Wright and Cassese, 1996) and
favoured the development of comparative policy research on policy-making in
European Member states and as part of the European integration process. Other
critics challenged this overly agency-driven approach to policy-making, and drew
on other social sciences traditions in order to develop alternative approaches
to policy studies, such as pragmatist and critical approaches to policy studies
(Laborier and Trom, 2003; Callon et al, 2013). These were particularly useful in
order to understand the changes underway at the public sphere’s margins and in
relationship with social innovations, conflicts and policy outsiders. Last but not
least, gender policy studies at an international level and in France highlighted
the benefits to be expected – methodologically, empirically and theoretically –
from applying a gender perspective to policy studies (Mazur and Revillard, 2016)
and more specifically, to the French concept of ‘policy analysis’ (Boussaguet and
Jacquot, 2009; Engeli and Perrier, 2015). By focusing on the role of women in
policy-making and policy studies, gender policy studies highlighted the distinctive
features of gender discriminations, alongside social and racial discriminations,
in recruitment processes in the political, the administrative and the academic
spheres (Achin and Bereni, 2013). Similar to the study of environmental policies,
the work done on gender equality policies challenged classic models about the
functioning of the state and policy-making in France, and led to alternative
conceptual and analytic frameworks that allowed the exploration of original forms
of policy-making, the growing interplay between state and non-state actors, and
innovative forms of policy instrumentation such as mainstreaming (Halpern and
Jacquot, 2015).
Together, these criticisms highlighted competing transformative dynamics
and demonstrated the limited explanatory power of the ‘French model of public
policy’ (Boussaguet et al, 2015). A clear shift towards a more sociological approach
was thereafter initiated, which implied the reference to the notion of ‘action
publique’ (literally, public action) (Commaille, 2014), that was considered more
neutral than that of ‘politiques publiques’ and enabled more emphasis to be put on
actors, institutions and dynamics both within and outside the state and public
organisations. A large consensus emerged in the 2000s among French scholars to
use the term ‘sociology of public action’ in order to label this academic field and to
reaffirm the specificity of the French perspective on policy studies: a sociological
perspective and a focus on public interventions (Boussaguet and Jacquot, 2010;
Boussaguet et al, 2015; see also Patrice Duran in Chapter Eighteen). This shift
was later confirmed in textbooks (Hassenteufel, 2011; Lascoumes et al, 2007)
and over the recent period, it has become commonly accepted that the study of
policy seeks, in a sociological perspective, to analyse the role of the state and its
restructuring, processes of government and governance, state–society relationships
and policies as both a process and an outcome by focusing on policymakers, their
practices, ideas and discourses, how they engage in sustained relationships with
a large variety of market and society actors, and the concrete devices they use
6
Introduction
in order to make policy objectives operational (Commaille and Jobert, 1999;

Commaille, 2014).
As a result, scholarly debates about policy studies in France have been influenced
since the 1980s onwards by, on the one hand, the above-mentioned specificity
and on the other hand, by successive imports of research and concepts that
were developed elsewhere, such as Charles Jones and Charles Lindblom during
the 1980s (Thoenig, 1985; Mény and, Thoenig, 1989), John Kingdon, Hugh
Heclo, Jeremy Richardson, Vincent Wright, Bob Jessop and Renate Mayntz
during the 1990s (Kingdon 2010; Jordan and Richardson, 1983; Heclo, 1974;
Wright, 1979; Jessop, 1996; Mayntz, 1975), Peter Hall, Frank Baumgartner, Paul
Pierson, Paul Sabatier, Claudio Radaelli or Vivien Schmidt during the 2000s7
(Hall and Taylor, 1996; Jones and Baumgartner, 2005; Pierson, 2004; Sabatier,
1998; Radaelli and Schmidt, 2015). Nevertheless, the contribution of French
scholars to international debates on public policy remained limited as shown by
the small number of references to French scholarly work in English textbooks,
apart from European studies (Saurugger, 2010).8 A first turn was identified in
the mid-2000s with the edited volume on policy instruments by Lascoumes and
Le Galès (2004; 2007; see also Halpern et al, 2014), which went beyond the
import of the policy tools’ approach that had been developed by Christopher
Hood and developed an original contribution to international debates on public
policy in both French and English (Margetts and Hood, 2016). Recently, the
pragmatist approach to policy making that was developed by Zittoun (Zittoun,
2014) in both French and English also contributed to international debates on
interpretative public policy analysis beyond the work by Fischer and Forrester
(Fischer and Gottweis, 2012).
As underlined by Jacques de Maillard and Andy Smith in Chapter Four, three
fundamental debates have structured and cleaved French policy studies throughout
the past three decades. The first concerns the role of ‘ideas’ in public policies,
the second the relationship between institutions and actors, while the third
is centred upon the role of the state and politics itself within policy-making,
implementation and evaluation. These controversies arose concomitantly to the
institutionalisation of the discipline as a field of scholarly work in its own right,
and shaped the demarcation of its boundaries in three different ways (see Chapter
Seventeen by Patrick Hassenteufel and Patrick Le Galès). The first way was by
engaging in theory building and seeking for its full integration in the political
science academic community, this led to excluding policy studies produced
outside academia such as policy evaluation, policy analysis, policy design, policy
management. The second way was through the streamlining of teaching content
as well as research design and methodologies used to study the policy process.
French policy studies, however, only selectively engaged with international debates
and controversies, although a greater cross-fertilisation between endogenous and
exogenous perspectives emerged in the late 1990s, alongside a greater willingness
to actively engage in international scientific fora. Over time this resulted in the
7
development of a number of policy studies subfields,9 which together, contribute

to shedding new light on policy analysis and studies in the French context.
This rapid overview helps us to understand the historically-situated process by
which the notion of ‘policy analysis’ was gradually replaced by that of ‘(political)
sociology of public action’, and why it is now understood in France as an academic
field of research and not as a practitioner’s field of knowledge. It also highlights
on the one hand, the lack of a problem-solving orientation of policy analysis in
the academic field, and on the other hand, why knowledge about policy analysis
in the policy process is being little addressed in scholarly work.
Policy analysis in France: an enduring disconnect between

academic and practitioners’ knowledge
The lack of a single, unified understanding, in the French context and beyond
the academic sphere, of what ‘policy analysis’ means and of ways to analyse it,
constituted both a semantic and an analytical challenge for the contributors to
this volume. Indeed, this had long-term consequences over the research focus and
availability of scholarly work about policy analysis. Any attempt to understand
the theory and practice of policy analysis requires a thorough exploration of the
distinction usually made in French academic and practitioner debates between
policy process studies and policy analysis. Rather than ignoring this challenge,
we deliberately considered it as a common starting point for questioning the
specificity of policy analysis in the French context, as well as an opportunity to
contribute to French and international debates among academics and practitioners.
In this section, we discuss the extent to which these semantic ambiguities
shaped the development of policy analysis in the French context in both academic
research and practitioners’ knowledge up to the present.
Looking for the forgotten problem-solving orientation
To begin with, we found that there were no notions or concepts available in

French in order to address problem-solving orientations. The process by which
concepts and ideas about policy analysis had been imported from the US and
hybridised with existing scholarly work in political science was largely dominated
by a sociological perspective (Chapter Four). Paradoxically, although academic
debates considered Charles Jones’ book An Introduction to the Study of Public Policy
a critical milestone in the French context, they ignored Jones’ call to shift from a
problem-solving orientation towards that of policy process studies (Jones, 1970).
In the case of public policy analysis, its early developments were dominated by a
strong sociological influence, and the work already underway in the 1970s as part
of the sociology of organisations and bureaucratic elites (Leca and Muller, 2008).
This led on the one hand to developing a comprehensive perspective on policy
processes, and on the other hand, to rejecting the problem-solving dimension of
policy analysis as being normative, marginalising the rational choice perspective
8
Introduction
and favouring qualitative sociological methods as argued by Claire Dupuy and

Philippe Zittoun in Chapter Five. Although the problem-solving orientation
repeatedly resurfaced during the early stages of policy studies, it was systematically
delegitimised by political scientists (Muller, 2015) and considered irreconcilable
with the subdiscipline’s strive for being formally acknowledged by political
science and sociology (Chapter Seventeen). The institutionalisation of policy
studies was strongly linked to the scholars’ ability to demonstrate their legitimacy
as academics by rejecting any policy analysis perspective (see above). These early
developments account for today’s weak relationships between academics and
practitioners. They also laid the groundwork for economists investing this area
of work (Chapter Sixteen).
Accounting for weak relationships between academics and practitioners
Unlike the situation observed in other countries, where such a relationship

between academics and practitioners would primarily derive from teaching and
training activities, the training of higher civil servants – and French elites in
general – is mainly organised outside universities, in the so-called Grandes Écoles,
which constitutes an education system of its own.10 In these schools, in which
both a generalist and a specialised knowledge is provided to the students who
successfully pass the entrance examination,11 a vast share of the courses are given
by alumni and rarely entail any research perspective. In the case of ENA – the
National School of Administration – the overrepresentation of public law and
general knowledge in the curriculum was progressively replaced, from the 1950s
onwards, by economics and a more specialised and case-oriented teaching (Biland
and Kolopp, 2013, 241–3). The content of entrance examinations followed a
similar path, and was progressively re-oriented towards policy problems and
technical case studies (Mangenot, 1999). At the same time, leading figures teaching
economics in a problem-solving and a ‘modernising’ perspective emerged within
the General Commission for Planning and Programming (Commissariat général
du Plan [CGP]) and the National Institute for Statistics and Economic Studies
(Institut national de la statistique et des études économiques [INSEE]), and were soon
considered major figures of the school (Kolopp, op cit). Mirroring the growing
divide between Grandes Écoles and the universities, the number of academics
involved in the formers’ administrative boards decreased steadily, while a quasi-
permanent educational team – mainly composed of civil servants – emerged
around the board (Biland and Kolopp, 2013).
Today’s weak relationships between academics and practitioners are also
explained in relation to the academics’ low contributions to policy developments
on issues of interest for the government and parliament. Unlike the situation
observed in other countries, where researchers are often engaged in the politico-
administrative system in order to produce commissioned work or to be consulted
on specific policy issues, this is seldom the case in the French context (Chapter
Eleven). When there is a need to address policy issues and mobilise expert
9
knowledge, the French bureaucracy usually tends to seek for policy analysis outside
academia and relies mostly on senior civil servants coming from the same state
elite network (Grands Corps), and as of late, on consultancy firms (Chapter Ten).
Interestingly, authors in this volume argue that parliament, party leadership and a
vast majority of local governments also rely upon within-state sources of expertise
and less so on academia. In this context, whether a cause or a consequence of the
historic divide between academics and practitioners, the work done by French
researchers rarely entails a problem-solving orientation that would facilitate their
contribution to political debates.
The meaning and practices of policy analysis in the French context
These past and recent evolutions are essential in order to fully grasp the meaning
and practices of policy analysis in the French context. Together, they account for
the development of two different dynamics in relationship with policy analysis,
which increasingly differ from one another and refer to highly differentiated
professional spheres: one within the academic field that primarily focuses on
policy process studies and rejects the idea of producing knowledge for policy
practice, and the other, which is developed by practitioners without reference
to academia rules and according to poorly defined methodologies. As a result,
and apart from gender policy analysis that seeks to systematically analyse policy
processes in a gender perspective,12 policy analysis remains both underdeveloped
as an output of academic research and understudied as an object of research.
While research on the role of actors in the policy process can easily be found
across scholarly research, there is little knowledge about policy analysis, how it is
produced and used during the policy process, and indeed, when embarking on
this book project, we found little systematic and up-to-date knowledge of how
and by whom policy analysis is produced outside the state, thereby requiring, in
the case of most chapters, the collection of new empirical evidence in order to
answer the question put to them as part of this volume.13 This constituted the
second major difficulty that we encountered in preparing this volume. Following
a North American perspective on policy analysis, each volume in this series
had to follow a similar outline in order to facilitate later comparative work but
proved complicated to follow in the case of this volume. Indeed, how does one
write a chapter on policy analysis in bureaucracies, parliament, political parties
and NGOs in the absence of any systematic empirical data and little information
provided in the reports produced by practitioners and the little research that
address this issue? Acknowledging this challenge, most authors considered this
book as an opportunity to develop a new perspective on the understanding of
policy analysis in France, and in some cases, it also opened fruitful perspectives
for future research.14
As a result, every chapter in this volume includes, on the one hand, an overview
of how policy studies account for the production and uses of policy analysis in a
given organisation, and on the other hand, they attempt to give a comprehensive
10
Introduction
view of policy analysis as produced and used by practitioners within and outside
the policy process. Interestingly, when considered as a policy resource likely to
be mobilised in the policy process in combination or in competition with others,
policy analysis offers a unique opportunity to assess and question the functioning of
policy-making in the French context as well as evolving state–society relationships.
This is developed in more detail in the following section.
The added value of policy analysis for understanding state

restructuring and policy developments in France
In this section, the emergence and strengthening of an autonomous field of state
expertise is explained in relationship with the role of state elites’ networks, the
centralisation of policy knowledge production and practices within the state
apparatus, and the limited attention given to policy analysis as part of non-state
and non-executive organisations’ strategies to increase their political capacity.
A long tradition of policy analysis inside the closed French bureaucratic system
Policy analysis, as an autonomous field of knowledge and expertise within the

state apparatus, developed from the eighteenth century onwards and, together
with other government tools and practices (Eymeri 2001; Foucault, 1978), it
contributed to the formation of the modern state and to making society ‘legible’
(Scott, 1998). The development of ‘governing knowledge’ is considered critical in
the strengthening of the central state before and after the French Revolution, as
pinpointed by Tocqueville in the Old Regime and the Revolution (1856), and strongly
related to the position enjoyed thereafter by state elites in the state apparatus and
policy-making. In Chapter Six, Emilie Biland and Natacha Gally argue that the
development of state elites’ networks – or Grands Corps that result either from
a generalist, administrative or from a specialised training in above-mentioned
Grandes Écoles as opposed to universities – constitutes an enduring mechanism
of selection and reproduction of senior civil servants. As such, it is an essential
component of the bureaucratic system’s stability over time and despite the changes
brought on by successive political regimes. Monopolising top positions within
and outside the state apparatus, including public- and private-owned enterprises,
state agencies, and research and innovation centres, these elite networks long had
a monopoly on the production and the selection of legitimate policy analysis
across policy areas. Considered one of the most visible dimensions of the so-
called ‘French model of public policy’ in the context of the Fifth Republic
(Muller, 2015), it contributed to the state centrality and to ensure a strong level
of autonomy in the choice and selection of policy priorities and solutions. To be
sure, such degrees of coherence should not be overestimated and this monopoly
was not exercised equivalently according to the Grands Corps themselves, to policy
areas and across society.15 This nevertheless holds several consequences for the
format, content and practices of policy analysis among practitioners.
11
Despite the Grand Corps’ effective weakening as a result of decentralisation,

privatisation and internationalisation dynamics in policy-making, most chapters
in this volume confirm the weakness of policy analysis outside the central state
apparatus – mainly central administrations – and state elites’ networks (see Chapter
Seven and beyond). In contrast with other countries, powerful ministerial cabinets
mainly staffed with higher civil servants are considered a specific feature of the
Fifth Republic’s politico-administrative system. Strategically positioned at the
intersection between policy-making and political action, they provide politicians
with policy knowledge emanating from departments, and conversely, they adapt
policy tools to the timing and constraints of political action (Eymeri, 2001;
2003). Also, the role played by these state elites’ networks partly accounts for
policy domains developing in silo and according to logics of their own, as well
as for some major differences between the form and content of policy analysis
depending on the Grand Corps and their respective power within the bureaucratic
system. Each central administration – or Ministry – generally have their own
inspectorate,16 in which senior civil servants can, at an advanced career stage,
potentially be asked to produce policy analysis on relevant policy issues on an
individual basis – most often a report – or as part of an ad hoc commission (see
Chapter Eleven). In addition, the enduring divide between the University and
the Grandes Écoles systems further increased each Grands Corps’ autonomy in the
selection, training and career evolutions of their members (see above). Insofar as
every member of a Grands Corps, and more generally, every higher civil servant
can rely upon her or his career and professional experience in order to produce
legitimate policy analysis, there is no single understanding among practitioners of
what policy analysis means. Unlike policy analysis in other national administrative
traditions or as produced in the private sector (for example, consultancy firm)
(see Chapter Ten), this critical policy resource is thereby characterised, in the
French context, by a lack of homogeneity and its overall weakness in terms of
both design and methodology. As a result, many authors in this volume struggled
to identify or produce a general and systematic overview of policy analysis as
produced by practitioners, and when considering the practices of producing
and assessing policy analysis, they were confronted with a general lack of clarity
and debate regarding the design and methods used as part of their inquiries (see
Chapter Five).
In addition to the robustness of state elites’ networks, the dominant role of the
executive branch constitutes a second historically embedded characteristic of
governing practices and state–society relationships in France.
The emergence and strengthening of an autonomous field of state expertise
Unlike other countries where policy knowledge primarily comes from academia
or think tanks, central administration – and the core executive branch – is regarded
historically as a privileged site for the production of knowledge pertaining to
the state and its interventions. Yet as discussed by Mathieu Hauchecorne and
12
Introduction
Etienne Penissat in Chapter Eleven, it was only during the post-war period
that a differentiated field of state expertise emerged in relationship with highly
differentiated reform agendas that nevertheless shared this common interest
in modernising social, economic and politico-administrative structures. This
centrality was durably reasserted – and strengthened following the introduction of
a semi-presidential Regime – the Fifth Republic – in 1958, thereafter leading to
the concentration of policy analysis inside the state apparatus despite subsequent
attempts to develop alternative sources of power. The enduring and dominant
role of the core executive branch in producing policy analysis and hierarchising
policy priorities during the policy process manifests itself in different ways.
As part of the modernist agenda to profoundly transform social, economic
and political structures, several attempts were made to break with traditional
bureaucratic hierarchies and career paths and to encourage the diffusion of
modernist ideas, practices and methods in the state apparatus (Cohen, 1998).
Alongside the development of planning and programming activities, administrative
task forces (administrations de mission) with extensive powers were introduced
and staffed with civil servants on secondment or non-civil servants on fixed-
term contracts. Among them, the General Commission for Planning (CGP)
was created in 1946 and rapidly endorsed a modernising role, by spreading out
macro-economic theories and defending a strong-willed conception of economic
policy and state intervention (Hayward and Watson, 1975). Reporting directly
to the Prime Minister, the CGP played a critical role in the development of
policy analysis across policy fields, and developed into a consultative arena in
which a large number of state and non-state organisations jointly developed five-
year plans that were submitted for approval by Parliament and enjoyed formal
legal force. Following De Gaulle’s return to power in 1958 and in a context of
economic growth, the role of CGP rapidly extended from those policy fields
considered strategic in the post-war reconstruction context (for example, charcoal,
electricity, agricultural machinery, transport infrastructures) towards modernising
the economy and society as a whole.
Alongside administrative task forces, departments within central administration
also developed their own policy knowledge – statistical data, performance
indicators, monitoring tools, and so on – either through the work done by
inspection bodies17 – both departmental and cross-departmental – or by
encouraging the creation of directorates dedicated to strategic policy planning.
Among them, the National Institute of Statistics and Economic Studies (INSEE),
that was created in 1946 as a general directorate of the Finance Department,
rapidly developed into a semi-autonomous knowledge-based organisation. While
it does not directly contribute to policy-making, the INSEE has long been
considered a key producer of statistical data on the French society, and plays a
critical role in shaping the categories through which public policies are formulated
and implemented (Desrosières, 1993). Ministerial Statistical Services (MSS) also
engage in the publication of sectorial knowledge likely to be mobilised in policy
formulation.18 The Finance Department’s powerful Division for Economic
13
and Financial Studies (Service des études économiques et financières) offers another
example of institutionalised sources of departmental policy knowledge. Created
in 1947 within the Treasury Directorate, it gained additional autonomy in 1965
when it developed into the Prospective Directorate (1965) and as such, a source
of reflexive knowledge over bureaucratic practices – including the selection of
cost and productivity indicators in the context of the Planning-Programming-
Budgeting System programme (Bezes, 2009, 68–71).
Moreover, a myriad of research and expert institutes affiliated with ministries
and the Prime Minister’s office also contribute to the production of statistical
knowledge, policy analysis, studies and reports that are likely to shape the
substance and format of public policies. Together, they constitute a prime source
of bureaucratic knowledge and policy recommendations intended to guide and
evaluate state intervention. As a result, and in the French context, a vast share
of practitioners’ knowledge for and about policy-making – which would be
considered the equivalent of ‘policy analysis’ in the United States – is produced in
these organisations. Despite repeated political attempts to foster alternative sources
of policy analysis through the creation and restructuring of bodies, councils and
task forces that would cut across bureaucratic boundaries and hierarchies, state
expertise enjoys a certain degree of autonomy. This has several consequences for
the understanding of the politics of policy analysis in the French context. First,
this explains why, on the one hand, the self-commissioning of administrative
reports and studies is considered common practice, while on the other hand,
politicians rely upon extensive cabinets in which political advisers must deal –
somewhat uneasily – with higher civil servants in order to rely upon alternative
sources of knowledge. Second, considering these research and expert institutes’
activities and production as mere policy knowledge would be a mistake. Insofar
as they are fully integrated to the production of state knowledge about the state
and specific policy areas, their activities and research outputs meet the state
needs and are shaped by within-state logics. Yet by mixing empirical description,
scientific objectives and political recommendations, they are fully integrated to
the state intervening processes which they are intended to analyse. As such, they
are considered a preferred source of analysis for scholars seeking to understand
the intellectual or symbolic frameworks of state intervention (Rhodes, 2011).
Third, the development of a field of state expertise played a critical role in the
history of French social and economic sciences, as it contributed to their continued
strengthening. In that it was not the output of imported academic knowledge,
state expertise in this field led to the emergence a new body of knowledge,19
that was indexed on public policy imperatives at first and subsequently gained
academic recognition as convincingly argued by Daniel Benamouzig and Frédéric
Lebaron in Chapter Sixteen. This relationship only reversed in recent years, with
state expertise not being considered a place of scientific innovation whereas both
economics and the social sciences have become institutionalised as scientific
disciplines in their own right in universities and research centres. During the post-
war period, economists actively engaged, within the state, in the production and
14
Introduction
analysis of public accounts, economic modelling and forecast. The creation and
rapid development of new institutes and administrative departments within the
ministry of Finance (Prospective Directorate, INSEE, and so on) justified the need
to develop specific training programmes within existing or new Grandes Écoles,
first the École polytechique, that trains civilian engineers and scientists, as well as
army officers, and second, the National School for Statistics and Economic Studies
(École nationale de la statistique et des études économiques [ENSAE]). At the same time,
economic departments inside universities gained additional salience and grew
more autonomous from law departments, thus opening new opportunities for
engaging in international research activities while at the same time maintaining a
certain level in mathematics and statistics. This allowed on the one hand publishing
in peer-reviewed journals and, on the other hand, developing strong links with
the tradition of engineers-economists as taught in the Grandes Écoles – although
the convergence only occurred in the 1980s and 1990s under the impulse of a
few academics20 and bureaucrats, such as Henry Guitton. Yet these developments
also induced a clear shift towards mainstream economics inside universities and
the CNRS, to the detriment of critical and heterodox conceptions, as well as
the political economy tradition close to the social sciences, which still remains
today, although now at the fringes of the discipline. This academic discipline’s
internationalisation and institutionalisation also induced some changes in the
work achieved by practitioners within the state apparatus in producing policy
knowledge. Academic work was more frequently referred to in the sphere of
public decision-making with some impact on public policies themselves that is,
on the one hand, the growing role of economic experts in budgetary, monetary
and economic performance policies, and on the other hand, the increasing use
of economic references in specific policy fields such as public health.
In the following section, we discuss the extent to which the focus on policy
analysis sheds new light on current restructuring dynamics in policy-making and
the state functioning.
Understanding changes in the politics and practices of policy

analysis in the French context
In line with recent academic work on the French state (Culpepper et al, 2006;
Hayward, 2007) and policy developments in France (Borraz and Guiraudon,
2008; 2011; Cole et al, 2013; Hassenteufel, Saurugger, forthcoming; Elgie et al,
2016) that emphasised the complex interplay between change and continuity,
we found that the production of policy analysis is still primarily achieved within
state organisations and by state elites. Historical developments of policy analysis
– the way it was produced, where it was located, and whether and by whom
it was strategically used – matter and help understanding the state’s continued
centrality in this process. Nevertheless, the politics of policy analysis, as well as the
power relations it entails, underwent some changes according to the strategic role
devoted to policy analysis as part of non-state and non-executive organisations’
15
strategies to increase their political capacity. Chapters in this volume discuss the
extent to which such transformations have taken place and account for resistances
and restructuring processes in the production and uses of policy knowledge both
within and outside the state.
The restructuring of the field of state expertise
Within the Executive branch, state elites and organisations resisted – with unequal
success – to repeated attempts to reduce their dominance by encouraging two
major developments that are not specific to the French context (Chapter Six).
First, the increasing influence of ‘numbers’ (quantitative analysis, statistics, and
nowadays, big data) compared to more traditional literary or legal skills. Second,
this was also achieved by shifting policy expertise downstream of the policy process,
as top civil servants are increasingly engaged in policy evaluation and performance
measurement. The ability of bureaucrats to master these forms of policy knowledge
conditions their ability to resist increased attempts by consultants, private-sector
experts and members of interest groups to challenge their leadership and emerge
as alternative and legitimate sources of knowledge on and for policies.In addition,
research services affiliated with the Prime Minister, including post-war planning
and programming committees, were profoundly transformed from 1995 onwards.
Some organisations were dismantled or transformed, while new ones were created
on a somewhat different format.21 In most cases, the general trend shows the
growing use of more diversified sources of policy knowledge, as exemplified in the
case of the Economic Analysis Council (Conseil d’Analyse d’Economique). Mainly
composed of university-based economists, its creation ended the monopoly
previously held over economic expertise by higher civil servants. Beyond this
specific example, the restructuring of state expertise is increasingly shaped by
institutionalising fields of academic knowledge in economics and social sciences
in the university system (Chapter Seventeen). This also contributes to redefining
power relations between higher civil servants and academics.
The transformations taking place within the field of state expertise also modify
the role of ad hoc commissions, to which the French government may resort to from
time to time, and the reports that they produce. The Prime Minister, sometimes
in cooperation with the President or a Minister, may indeed prefer to resort to
ad hoc commissions, which are considered more flexible than aforementioned
research organisations and committees. Members’ selection offers more room
for manoeuvre and shows the tendency to draw on diversified sources of policy
knowledge, including academics. Insofar as their missions are closely related to
governmental strategies, these organisations have a greater capacity to produce a
more politicised analysis in their reports and practical recommendations.22
16
Introduction
Unequal policy knowledge capacity outside the politico-administrative system
Yet other organisations across levels of government and the public, the private
and the voluntary sectors, have developed their own capacity for analysing policy
processes. Whether political institutions (the Legislative branch, subnational levels
of government) or non-state organisations (consultants, think tanks, interest
groups, NGO’s, academics), policy analysis is not mobilised in a straightforward
way (Chapters Seven to Fifteen).23 As discussed in the following paragraphs, this
is not only explained by state organisations and elites resisting the development
of alternative sources of policy knowledge but also due to the limited interest
in developing policy analysis resources as part of non-state and non-executive
organisations’ strategies to increase their political capacity.
Despite successive waves of decentralisation from 1982 onwards, subnational
levels of governments have unevenly prioritised the need to develop their own
capacity for producing and evaluating public policies, as argued in Chapter Seven
by Alain Faure and Emmanuel Négrier (see also Loughlin, 2007; Galès, 2011;
Pasquier 2012; Pasquier et al, 2011). Most levels of government have gained
considerable powers together with the financial and administrative means to
endorse them. In addition, restructuring forms of local political leadership also
show the ability of local authorities and politicians to develop alternative political
resources in order to expand their political capacity vis-à-vis the French state
(Pinson, 2009). Yet when considering policy analysis itself, only a limited number
of large cities, together with some regions and counties have systematically
developed their own capacity for policy knowledge in close partnership with
consultancy agencies, mixed economy systems and European and International
organisations (Cadiou, 2005). A vast majority of subnational authorities still rely
heavily on the central administration’s expertise al administrationonal authorities
still or policy knowledge in close partnership with consultancy interventionism
in shaping policy developments at subnational levels of government is less related
to its own representatives and resources, including in terms of policy knowledge,
but rather to the diffusion of standards, best practices and ideas that allow some
level of territorial differentiation in combination with standardisation processes
(Douillet et al, 2012).
In the case of parliament, power asymmetry between the executive and the
legislative branches partly account for the limited development of policy analysis,
and several explanations are brought forward. As shown in Chapter Eight by
Olivier Rozenberg and Yves Surel, the French parliament is counted among
the weakest legislative institutions across western democracies when it comes
to assessing its role in the decision-making process. Moreover, not only are
there limited incentives for developing policy knowledge resources within the
parliament, but considering the limited MPs’ ability to influence policy-making,
they are less induced to specialise in specific policy issues and to develop policy
knowledge across policy domains.24 As a result, parliamentary debates tend to
be more political than technical, and rarely draw on the parliament’s own policy
17
knowledge resources. Despite several attempts to non-legislative activities,

including policy evaluation, the Constitutional Council systematically maintained
a restrictive approach of parliament’s activities in its ruling. To the exception of
the work achieved under the leadership of the national Assembly’s Commission
for the Assessment and Monitoring of Public Policies, which was created in 2009,
the greater part of policy evaluation activities remains thereafter undertaken within
government bureaucracy.25
The lack of incentives for developing policy analysis resources is further
exacerbated in the case of non-state actors, and sometimes explained by high
levels of interrelationships with higher civil servants. In Chapter Twelve, Camilo
Argibay, Rafaël Cos and Anne-Cécile Douillet highlight the paradox between, on
the one hand, the institutionalisation since the 1970s onwards of party-based policy
analysis resources, but on the other hand, their structurally unstable and fragile
positions – mostly due to within-party competition. Moreover, they enjoy little
autonomy from central administration, as party leadership favours contributions
from senior civil servants with whom they often share similar training. Often
serving in the most prestigious departments, they supply party leadership with
policy briefs and directly applicable recommendations. This confirms strong levels
of mutual acquaintanceship between governing parties and state organisations.
In a similar vein, Julie Gervais and Frédéric Pierru (Chapter Ten) argue that
many consultants and higher civil servants share common interests and values,
and follow similar careers paths. Up until the 1990s, policy knowledge focusing
on modernising public services and the state was jointly produced by higher civil
servants, political elites, some business leaders, alongside a few carefully selected
academics, intellectuals, journalists and union leaders gathered in closed clubs.
Following the diffusion of new public management (NPM) ideas and practices
across continental Europe, the influence of big international consultancy firms
– mostly Anglo-American – grew rapidly as a vast share of publically-owned
enterprises underwent privatisation reforms. Contacts with senior civil servants
intensified (Bezes, 2012) through these consultancy firms’ increased interest in
government activities, and their ability to attract higher civil servants and graduates
from the Grandes Écoles. A hybrid type of elite emerged and thereafter actively
engaged in the diffusion of NPM ideas in the French state.
In the case of think tanks, however, also discussed in Chapter Twelve, and by
contrast with their British or American counterparts, these organisations enjoy
limited financial and human resources. Even though they actively engage in the
production and diffusion of policy knowledge and ideas, their political role is
considerably less institutionalised (Medvetz, 2012).
Another explanation for remaining asymmetries between state and non-state
actors in controlling policy knowledge resources lies in the strength of the belief
– inherited from the French Revolution – that the state best embodies and
guarantees the general interest. Historically, this long constituted a major obstacle
to the development of civil society organisations, including non-governmental
organisations, economic interest groups and to a lesser extent trade unions, and
18
Introduction
often justified their limited interest in developing policy knowledge resources

as opposed to other, more effective, influence strategies. By focusing on the
production of policy knowledge, contributors to this volume, to a degree, confirm
that such an analysis holds true up until the 1990s for these various groups, but
show how, over the recent period, they increasingly engage in such activities with
a limited impact on policy formulation as opposed to policy implementation.
In the case of business organisations, Guillaume Courty and Marc Milet, in
Chapter Thirteen, show how successive governments entrusted these groups
with the role of information and socio-economic data providers. They actively
engaged in the development of statistics – even though state agents successfully
claimed and monopolised this policy resource since the post-war period – and
fully committed to transmitting information and figures that civil servants became
accustomed to receiving from them. As of late, liberalisation policies, state reforms
and the formalisation of lobbying activities, somewhat transformed relationships
between business organisations and government organisations. To begin with,
a large number of the consultative arenas that dated back to the post-war era
were dismantled or restructured, while at the same time, the government’s
dependency on business organisations’ data and policy knowledge decreased. In
addition, business organisations now increasingly rely on outsourcing, mainly
to consultancy firms, for the production of quantitative data through surveys,
but when interacting with public authorities (for example, working groups or
parliamentary hearings), their representatives still favour policy resources produced
internally.
Similarly, Sophie Béroud and Jean-Marie Pernot in Chapter Fifteen account for
the various ways in which trade unions selectively developed policy knowledge
resources and made continued efforts to analyse their socio-economic environment
– even on a small-scale and ad hoc basis. Together with employers’ organisations,
trade unions jointly managed whole segments of the welfare state (for example,
health insurance, retirement pensions, unemployment insurance and job training),
thereafter developing considerable policy knowledge in these areas despite the
state retaining the final say in defining policy orientations. Nevertheless, their
influence on policy-making has been strongest during exceptional times, such
as during the Second World War, or May 1968 events, which confirms their
successful ability to establish a favourable balance of power by drawing on protest
rather than mobilising policy knowledge during consultation and negotiation
arenas. Since the 2000s, however, new forms of social dialogue between employers
and trade unions have been introduced with the support of the state, thus leading
these organisations to increasingly consider the production of policy knowledge
resources as critical in their attempts to influence the formulation of labour
policies.
In the case of non-governmental (NGOs) and civil-society organisations
(CSOs), Laurie Boussaguet and Charlotte Halpern (Chapter Fourteen) argue that
both groups have increased their policy knowledge resources and increasingly
consider it as a strategic action repertoire. Notwithstanding some variations, these
19
groups were categorised – and to some extent still are – as policy outsiders and
remained largely secluded from policy formulation or in some cases, selectively
included at policy implementation stages. As a result, they had little incentive to
include policy knowledge resources in developing their own strategies unless it
contributed to establishing a favourable balance of power during protest cycles
and as part of court appeals, or to access public funding and subsidies. Recent
evolutions show, however, that CSOs and NGOs across policy fields and levels
of government cannot be considered any more as mere service providers or
protesters, as they increasingly engage in the production of policy knowledge and
seek for formal recognition as policy stakeholders. Meanwhile, state organisations
and elites increased their capacity to internalise civil society organisations’ policy
knowledge in the policy process.
In Chapter Nine, Cécile Blatrix and Guillaume Gourgues further discuss the
added value of public participation devices to the development of alternative
policy knowledge resources. Similar to other cases considered in this book
series, participatory devices were introduced in a variety of formats (public
inquiries, users’ committees, public hearings, consensus conferences, participatory
budgeting, and so on) since the 1990s onwards. While they seldom led to the
production of alternative policy knowledge resources, these devices made it
possible for decision-makers in the political or administrative spheres to selectively
enrol knowledge, opinions and actors from outside the state apparatus, thus in
some cases contributing to the cognitive reframing of specific issues. Under such
conditions, dominant policy frames were adapted and, more rarely, gave rise
to new forms of policy knowledge. Participatory devices also helped making
certain areas of knowledge and dimensions of a given policy issue visible which
had been hitherto hardly discernible. From time to time, they play a role in the
adaptation, circulation and exchange of knowledge produced by a vast share of
actors, including interest groups, local residents and professional experts.
Together, chapters in this volume shed new light on policy analysis in the
French context – how and by whom it is produced, its contents, forms and uses,
as well as its long-term impact on state–society relationships and policy dynamics.
It provides some explanation for the enduring gap between knowledge for and
of policies, and as such, an opportunity to question the added value of policy
analysis as an object of study and a government practice.
Book outline
The book is divided into five sections and 18 chapters. The first part of the
book focuses on the styles and methods of public policy analysis. The first three
chapters provide a history of the development of policy analysis in France. Three
distinct periods are examined: first, the prehistory from the nineteenth century
until the 1970s; second, the invention of policy analysis in the 1970s and 1980s;
and, third, the controversies and debates that have shaped policy analysis over the
20
Introduction
past 15 years. A fourth chapter completes this history by focusing on the various
methods and research designs that are used in this academic field.
The second part of the book examines policy analysis in government or, in
other words, the study for policy and production of knowledge meant as a way
to improve policy-making and outcomes. It is divided into three chapters, each
of them highlighting the importance of knowledge production for policy within
the state apparatus, local governments, and the parliament. This second part pays
particular attention to the historical and current role played by civil servants and
more specifically the ‘Grands Corps’ in this process. In all three institutional arenas,
this specific group of elites has enjoyed differentiated degrees of autonomy in
producing knowledge and expertise, as well as in influencing rulemaking.
Part Three of the volume includes chapters that consider the role of committees,
public inquiries, and consultants in the public policy process. In other words,
this third part focuses on specific arenas of political participation that were
artificially created by the state in order to consult and include organised civil
society organisations either on a permanent or an ad hoc basis. While the French
political system is unquestionably characterised by the leading role played by
civil servants within policy-making, the development of committees and public
inquiries, together with the continued presence of consultants and councillors,
also constitutes another feature of policy-making and knowledge production
in policy processes in the French context. These chapters discuss the extent to
which these groups and individuals have come to exert a growing influence in
recent years.
Part Four focuses on the study for policy and knowledge production within
political parties, think-tanks and interest groups. In many ways, the French state
remains characterised by a weaker tradition than that of negotiation in Germany
and that of pluralism of influence (for example, think tanks and interest groups) in
the United States. These three chapters suggest exploring the French specificity by
highlighting the large array of action repertoires and influence strategies that are
used by political parties and interest groups in order to influence policy-making.
Although knowledge production for policy remains underdeveloped vis-à-vis the
situation observed in other western democracies, it has played a growing role
over the past 15 years both due to internal factors as well as the state’s initiative.
Part Five of the book critically assesses the place of policy analysis within the
academic world. It is divided into three different chapters. The first chapter shows
how economists and their specific understanding of policy analysis have come to
occupy the vacuum lying between the study of policy and the study for policy.
The two remaining chapters critically reflect on current relationships between
the public policy academic community and places of knowledge production for
policy within the state. This helps to explain why the distinction between policy
analysis and policy studies still holds.
21
Notes
1 We are grateful to Iris Geva-May and Michael Howlett, the editors of the International Library
of Policy Analysis, for their support throughout the preparation of the volume and for that of
Monica Steffen in launching the project.
2 Additional cross-cutting perspectives, such as enduring socio-spatial inequalities in the
recruitment of French elites or a gender perspective on the study for and of policy, have been
systematically included in all chapters where appropriate.
3 We had two opportunities to meet as a group during the preparation of the book. First at a
workshop hosted by the Centre d’études européennes at Sciences Po, in Paris in April 2015
and second, at a panel during the ICPP Conference in Milan in July 2015.
4 In the first translation of Lasswell, Aron and Lerner’s book, The Policy Sciences in the United
States, translated as Les sciences de la politique aux Etats-Unis.
5 It should be noted that politics, policy and polity are commonly referred to, in French, by using
the word ‘policy’. Moreover, by adding the word ‘publique’, policies are implicitly considered
public.
6 Two founding moments that both took place in the early 1980s were particularly critical: first,
the organisation of a dedicated round table as part of the first national congress of the French
political science association (AFSP) in 1981, and second, the introduction of a dedicated volume
in the First French Political Science Treaty (Thoenig, 1985).
7 See also the recently published Oxford Handbook of French Politics which discusses the specificity
of policy developments in the French context as well as the specific contribution of policy
studies as developed in France (Elgie et al, 2016).
8 Even though it provides a comparative connection, the contribution of French gender policy
studies also remains distinct (Mazur and Revillard, 2016).
9 Similar to the situation observed at the international level, gender policy studies is particularly
representative of this recent development, with two reference books having been produced
(Jacquot and Mazur, 2004; Achin and Bereni, 2013) together with a number of published pieces
on gender policy studies in France and in a comparative perspective (Mazur and Revillard,
2016).
10 Albeit with some minor differences, this has been the case since the late nineteenth century.
11 Barriers explanatory of unequal access for specific social groups and women to the Grandes Écoles
and to top positions in the administrative and the political spheres are discussed in a number
of chapters in the book. For example, gender policy studies identified the barriers faced in
promoting a feminist critical analysis of policy issues or in breaking the glass ceiling in order
to access policy-making in so called ‘sovereign’ policy domains as opposed to social and health
policies (Revillard, 2009, 44). See also Chapter Six by Biland and Gally in this volume.
12 This refers to the policy practitioner role within the women’s policy agencies since the 1960s
and their links to the field of ‘women’s advocacy’ (Bereni, 2009) in the French context. Beyond
women’s representation, the creation of specialised observatories contributed to the production
of information and policy recommendations about cross-sectional or emerging issues (parity,
ethnic discriminations, and so on) in support of new policy areas or measures.
13 It should be noted that a strong research tradition developed, both in political science and
sociology, in line with Foucault’s work on practices of governmentality (Foucault, 1978) in
order to account for such early developments. See in particular the work done on cameral
sciences (Laborier et al, 2011) and sciences of government (savoirs de gouvernement) (Ihl et al,
2003; Kaluszynski and Payre, 2013) in a historical perspective. Yet it is only in the recent period,
with the work initiated by Lascoumes on policy instrumentation (Lascoumes, 2003; Lascoumes
and Le Galès, 2004), that its implication for policy process studies were highlighted (Halpern
et al, 2014).
14 Based on the perspective structuring of this book the three editors organised a panel on this
topic at the 2017 Congress of the French Political Science Association.
22
Introduction
15 For example, it is impossible to understand past and recent developments across policy areas
such as energy, transport, telecommunications and industry without taking into account the
central role played by two Grands Corps – École des Mines and École des Ponts and Chaussées –
both within and outside the state apparatus, including public and private owned enterprises,
and in specialised state agencies that were created at the local level in order to run strategic
services (for example water, health, or housing). Similarly, state elites who were trained as
generalists (Inspection des finances, Conseil d’Etat, Cour des comptes, and so on) hold similar
positions in the state apparatus, administrative and fiscal jurisdictions, and, at subnational levels
of government, as the state’s representatives (Préfets).
16 Those who do not, such as environmental affairs, usually count among the weakest in the state
apparatus (Lascoumes, 2012).
17 In the case of classified installations inspectorates, see Bonnaud (2005).
18 Among them, the Directorate for research, studies and statistics (DARES) at the Department
of Work and Employment, and the Directorate for research, studies, evaluation and statistics
(DREES) at the Department of Social Affairs.
19 This was particularly the case in economics, thus justifying the choice made by this volume’s
editors to depart from the series’ recommended outline in order to devote a specific chapter to
this issue.
20 Within the French Association of Economics and following the arrival of a new generation of
academics in major universities like the Sorbonne (Paris-1) or Nanterre.
21 Among others, the Economic Analysis Council (Conseil d’Analyse Economique) in 1997, the
Pension Orientation Council (Conseil d’Orientation des Retraites) in 2000, and the Employment
Orientation Council (Conseil d’Orientation pour l’Emploi) in 2005.
22 See the example of the Commission on the Measure of Economic and Social Performance,
created in 2008 on the initiative of President Sarkozy. It was chaired by three economists (Jean-
Paul Fitoussi, Amartya Sen and Joseph Stiglitz) and composed entirely of academics.
23 In all these chapters, where appropriate, some attention was given to above-mentioned cross-
cutting perspectives to the study for and of policy.
24 This is all the truer that parliament only has a limited number of standing committees, namely
eight in the National Assembly since 2009 for 577 MPs. By contrast, the US Senate has 16
standing committees and 88 committees for 100 senators.
25 In contrast with the way the US Congress maintains political control over the Executive branch
by mobilising policy analysis resources with the support from independent agencies such as the
Government Accountability Office.
References
Achin, C., Bereni, L. (eds) (2013) Dictionnaire genre and science politique: Concepts,
objets, problèmes, pp. 13–41, Paris: Presses de Science Po.
Anderson, J.E. (1979) Public Policy-Making, Austin, TX: Holt, Rinehart and
Winston.
Bardach, E. (2008) A Practical Guide for Policy Analysis: The Eightfold Path to More
Effective Problem Solving (4th edn), Los Angeles, Thousand Oaks: Sage, CQ Press.
Bereni, L. (2009) ‘Quand la mise à l’agenda ravive les mobilisations féministes:
L’espace de la cause des femmes et la parité politique (1997–2000)’, Revue
Française de Science politique 59(2), 301–23.
Bezes, P. (2009) Réinventer l’Etat: Les réformes de l’administration française (1962–
2008), Paris: Presses universitaires de France.
23
Bezes, P. (2012) ‘Etat, experts et savoirs néo-managériaux : Les producteurs et

diffuseurs du New Public Management en France depuis les ann.es 1970’, Actes
de la recherche en sciences sociales 193, 16–37.
Biland, E., Kolopp, S. (2013), ‘La fabrique de la pensée d’État’, Gouvernement et
action publique 2(2), 221–48.
Bonnaud, L. (2005) ‘Au nom de la loi et de la technique: l’évolution de la figure
de l’inspecteur des installations classées depuis les années 1970’, Politix, 24(69),
131–61.
Borraz, O., Guiraudon, V. (eds) (2008) Politiques publiques 1, Paris: Presses de
Sciences Po.
Borraz, O., Guiraudon, V. (eds) (2011) Politiques publiques 2, Paris; Presses de
Sciences Po.
Boussaguet, L., Jacquot, S. (2009) ‘Les trois âges du genre’, Revue Française de
Science Politique 59(1), 173–81.
Boussaguet, L., Jacquot, S. (2014) ‘Que nous enseignent les manuels de politiques
publiques?’, Revue française de science politique 60(6), 1175–8.
Boussaguet, L., Jacquot, S., Ravinet, P. (2014) Dictionnaire des politiques publiques,
Paris: Presses de Sciences Po (4th edn).
Boussaguet, L., Jacquot, S., Ravinet, P. (2015) Une ‘French touch’ dans l’analyse
des politiques publiques?, Domaine Gouvernances, Paris: Presses de la Fondation
nationale des sciences politiques.
Cadiou, S. (2005) ‘Vers une action urbaine “moderniste”: Les effets du discours
des experts savants’, Sciences de la société 65, 9–27.
Callon, M., Lascoumes P. and Barthe, Y. (2013) Acting in an uncertain world,
Boston, MA: MIT Press.
Cohen, A. (1998), ‘Le Plan Schuman de Paul Reuter’, Revue Française de Science
Politique, 48(5), 645–63.
Cole, A., Meunier, S., Tiberj, V. (eds) (2013) Developments in French politics 5,
Basingstoke, New York: Palgrave Macmillan.
Commaille, J. (2014) ‘Sociologie de l’action publique’, in L. Boussaguet, S.
Jacquot, P. Ravinet (eds) Dictionnaire des politiques publiques, pp. 599-607, Paris:
Presses de Sciences Po, 4th edn.
Commaille, J., Jobert, B. (1999) Les métamorphoses de la régulation politique, Droit
et société 24, Paris: Librairie Générale de Droit et de Jurisprudence (LGDJ).
Crozier, M., Friedberg, E. (1977) L’acteur et le système, Paris: Seuil.
Crozier, M., Thoenig, J.-C. (1975) ‘La régulation des systèmes organisés
complexes: le cas du système de décision politico-administratif local en France’,
Revue française de sociologie, 3–32.
Culpepper, P.D., Hall, P.A., Palier, B. (eds) (2006) Changing France: The politics
that markets make, Basingstoke, New York: Palgrave Macmillan.
De Beaumont, G., De Tocqueville, A. (1845) Système pénitentiaire aux États-Unis
et de son application en France: suivi d’un appendice sur les colonies pénales, et de notes
statistiques, C. Gosselin, Paris: Hachette livre BNF.
Desrosières, A. (1993) La politique des grands nombres, Paris: La Découverte.
24
Introduction
Dobuzinskis, L., Howlett, M., Laycock, D.H. (2007) Policy Analysis in Canada:
The State of the Art, Toronto: University of Toronto Press.
Douillet, A.-C., Faure, A., Halpern, C., Leresche, J.-P. (2012) L’action publique
locale dans tous ses états: Différenciation et standardisation, Logiques politiques,
Paris: l’Harmattan.
Dupuy, F. and Thoenig, J.C. (1983) Sociologie de l’administration française, Paris:
Armand Colin.
Elgie, R., Grossman, E., Mazur, A. (eds) (2016) The Oxford Handbook of French
Politics, Oxford: Oxford University Press.
Engeli, I., Perrier, G. (2015) ‘Pourquoi les politiques publiques ont toutes quelque
chose en elles de très genré’, in L. Boussaguet, S. Jacquot, P. Ravinet (eds) Une
‘French touch’ dans l’analyse des politiques publiques?, pp. 349–75, Paris: Presses
de Sciences Po.
Eymeri, J.-M. (2001) La fabrique des énarques, Études politiques, Paris: Economica.
Eymeri, J.-M. (2003) ‘“Frontière ou marches ? De la contribution de la haute
administration à la production du politique”’, in J. Lagroye (ed) La politisation,
pp. 47-77, Paris: Belin.
Favre, P. (1989) Naissance de la science politique en France, 1870–1914, Paris: Fayard.
Fischer, F., Gottweis, H. (2012) The Argumentative Turn Revisited: Public Policy as
Communicative Practice, Durham, NC: Duke University Press.
Foucault, M. (1978) Sécurité, territoire, population : cours au Collège de France (1977-
1978), Paris: Gallimard Seuil, 2004.
Gaudin, J.-P. (2007) Gouverner par contrat (2nd edn), Paris: Presses de Sciences Po.
Grawitz, M., Leca, J., Thœnig, J.-C. (1985) Traité de science politique, Paris: Presses
universitaires de France.
Hall, P.A., Taylor, R.C.R. (1996) ‘Political science and the three new
institutionalisms’, Political Studies 44(5), 936–57.
Halpern C., Jacquot, S. (2015) ‘Quelle actualité de la notion de secteur?’, in
L. Boussaguet, S. Jacquot, P. Ravinet (eds) Une ‘French touch’ dans l’analyse des
politiques publiques?, Paris: Presses de Sciences Po.
Halpern, C., Lascoumes, P., Le Galès, P. (eds) (2014) L’instrumentation de l’action
publique, Paris: Presses de Sciences Po.
Hassenteufel, P. (2008) Sociologie politique: l’action publique, Paris: Armand Colin,
2011.
Hassenteufel, P. (2011) Sociologie politique: l’action publique (2nd edn), Paris:
Armand Colin.
Hassenteufel, P., Saurugger, S. (forthcoming) Politiques publiques 4, Paris: Presses
de Sciences Po.
Hassenteufel, P., Zittoun, P. (2017) ‘From policy analytical styles to policymaking
styles’, in M. Brans, I. Geva-May, M. Howlett, Handbook of Comparative Policy
Analysis, Abingdon: Routledge.
Hayward, J. (2007) Fragmented France: Two centuries of Disputed Identity, Oxford:
Oxford University Press.
25
Hayward, J., Watson, M. (eds) (1975) Planning, politics and public policy: The British,
French and Italian experience, London: Cambridge University Press.
Heclo, H. (1974) The Private Government of Public Money: Community and Policy
inside British Politics, Basingstoke: Macmillan.
Ihl, O., Kaluszynski, M., Pollet, G. (eds) (2003) Les sciences de gouvernement, Paris:
Economica.
Jacquot, S., Mazur, A. (2004) ‘Politiques publiques et genre’, in L. Boussaguet,
S. Jacquot, P. Ravinet (eds) Dictionnaire des politiques publiques, Paris: Presses de
Sciences Po (4th edn), 2010, pp. 460–9.
Jessop, B. (1996) ‘Post-Fordism and the state’, in B. Greve (ed) Comparative Welfare
Systems, pp. 165–83, Basingstoke: Palgrave Macmillan.
Jobert, B. (1994) Le tournant néo-libéral en Europe: idées et recettes dans les pratiques
gouvernementales. Logiques politiques 21, Paris: l’Harmattan.
Jobert, B., Muller, P. (1987) L’Etat en action: Politiques publiques et corporatismes,
Paris: Presses universitaires de France.
Jones, B.D., Baumgartner, F.R. (2005) The Politics of Attention: How Government
Prioritizes Problems (new edn) Chicago, IL: University of Chicago Press.
Jones, C.O. (1970) An Introduction to the Study of Public Policy, Belmont, CA:
Wadsworth.
Jordan, A.G., Richardson, J.J. (1983) ‘Policy communities: The British and
European policy style’, Policy Studies Journal 11(4), 603–15.
Kaluszynski, M., Payre, R. (eds) (2013) Savoirs de gouvernement: circulation(s),
traduction(s), réception(s), Paris: Economica.
Kingdon, J.W. (2010) Agendas, Alternatives, and Public Policies (2nd edn), London:
Longman.
Kolopp, S. (2013) ‘De la modernisation à la raison économique: La formation en
économie à l’ENA et les déplacements des lieux communs de l’action publique
(1945–1984)’, Genèses 93(4), 53–75.
Laborier, P., Trom, D. (2003) Historicités de l’action publique: [actes du colloque,
Amiens, 12–13 octobre 2000], Paris: Presses universitaires de France.
Laborier P., Audren F., Napoli P. and Vogel J. (dir) (2011) Les Sciences camérales,
activités pratiques et histoire des dispositifs publics, Paris: PUF & CURAPP.
Lascoumes, P. (2003) ‘Gouverner par les instruments, ou comment s’instrumente
l’action publique?’, in J. Lagroye, La Politisation, Paris: Belin.
Lascoumes, P. (2012) Action publique et environnement, Paris: PUF.
Lascoumes, P., Le Galès, P. (eds) (2004) Gouverner par les instruments, Paris: Presses
de Sciences Po.
Lascoumes, P., Le Galès, P. (2011) Sociologie de l’action publique, Paris: Armand
Colin, 2nd edn.
Lasswell, H.D. (1951) The Policy Sciences: Recent Developments in Scope and Method,
Stanford, CA: Stanford University Press.
Lasswell, H.D., Aron, R., Lerner, D. (1951) Les ‘Sciences de la politique’ aux
Etats-Unis, Cahiers de la Fondation nationale des sciences politiques 19. Paris:
Armand Colin.
26
Introduction
Le Galès, P. (2011) Le retour des villes européennes: Sociétés urbaines, mondialisation,

gouvernement et gouvernance (2nd edn), Paris: Presses de Sciences Po.
Leca, J., Muller, P. (2008) ‘Y a-t-il une approche française des politiques publiques?
Retour sur les conditions de l’introduction de l’analyse des politiques publiques
en France’, in P. Warin, O. Giraud (eds) Politiques publiques et démocratie, pp.
35–72, Paris: La Découverte.
Leca, J., Favre, P., Hayward, J.E.S., Schemeil, Y. (2003) Être gouverné: études en
l’honneur de Jean Leca, Paris: Presses de Sciences Po.
Lindblom, C.E., Woodhouse, E.J. (1992) The Policy Making Process (3rd edn),
London: Pearson.
Loughlin, J. (2007) Subnational Government: The French experience, Basingstoke,
New York: Palgrave Macmillan.
Maillard, J. de, Kuebler, D. (2009) Analyser les politiques publiques, Fontaine: Presses
universitaires de Grenoble, 2014. Majone, G. (1992) Evidence, Argument and
Persuasion in the Policy Process, New Haven, CT: Yale University Press.
Mangenot, M. (1999) ‘‘‘L’entrée en technocratie.” Le concours de l’ENA et les
transformations du modèle du haut fonctionnaire’, in D. Dulong, V. Dubois
(eds) La question technocratique. De l’invention d’une figure aux transformations de
l’action publique, pp. 93-107, Strasbourg: Presses Universitaires de Strasbourg.
Margetts, H., Hood, C. (2016) ‘Tools approaches’, in Contemporary Approaches to
Public Policy, pp.133–54, New York: Springer.
Mayntz, R. (1975) Policy-making in the German Federal Bureaucracy, Amsterdam
and New York: Elsevier.
Mazur, A., Revillard, A. (2016) ‘Gender policy studies: Distinct, but making the
comparative connection’, in R. Elgie, E. Grossman, A. Mazur (eds) The Oxford
Handbook of French Politics, pp. 556–82, Oxford: Oxford University Press.
Medvetz, T. (2012) Think Tanks in America, Chicago, IL: University of Chicago
Press.
Mény, T. (1989) Les politiques publiques, Paris: Presses universitaires de France.
Mény, Y., Thœnig, J.-C. (1989) Politiques publiques, Thémis, Paris: Presses
Muller, P. (2015) Politiques publiques, 11th edn, Paris: Presses Universitaires de
France.
Muller, P., Surel, Y. (2000) L’Analyse des politiques publiques, Paris: Montchrestien,
Padioleau, J.-G. (1982) L’État au concret, Sociologies 20, Paris: Presses universitaires
de France.
Pasquier, R. (2012) Le pouvoir régional: Mobilisations, décentralisation et gouvernance
en France, Paris: Presses de Sciences Po.
Pasquier, R., Guigner, S., Cole, A. (2011) Dictionnaire des politiques territoriales,
Paris: Presses de Sciences Po.
Pierson, P. (2004) Politics in Time: History, Institutions, and Social Analysis, Princeton,
NJ: Princeton University Press.
Pinson, G. (2009) Gouverner la ville par projet: Urbanisme et gouvernance des villes
européennes, Paris: Presses de Sciences Po.
27
Radaelli, C.M., Schmidt, V. (2015) Policy Change and Discourse in Europe,

Abingdon: Routledge.
Revillard, A. (2009) ‘Le comité du travail féminin et la genèse d’une politique
d’égalité professionnelle en France (1965–1983)’, Revue Française de Science
Politique 59(2), 279–300.
Rhodes, R.A.W. (2011) Everyday life in British government, Oxford: Oxford
University Press.
Sabatier, P.A. (1998) Theories of the Policy Process, Boulder, CO: Westview Press Inc.
Saurugger, S. (2010) Théories et concepts de l’intégration européenne, Paris: Presses
de Sciences Po.
Scott, J. (1998) Seeing like a State, New Haven and London: Yale University Press.
Thoenig, J.-C. (1985) ‘L’analyse des politiques publiques’, in M. Grawitz, J.
Leca (eds) Traité de science politique, vol 4, pp. 1–60, Paris: Presses universitaires
de France.
Tocqueville, A. (1856) Old Regime and the Revolution, Chicago: Chicago University
Press, 1998-2001.
Wildavsky, A. (1987) Speaking Truth to Power: The Art and Craft of Policy Analysis,
Piscataway, NJ: Transaction Publishers.
Wright, V. (1979) Conflict and Consensus in France, London: F. Cass.
Wright, V., Cassese, S. (1996) La recomposition de l’État en Europe, Collection
Recherches, Paris: Éd. la Découverte.
Zittoun, P. (2014) The Political Process of Policymaking: A Pragmatic Approach to
Public Policy, Basingstoke: Palgrave Macmillan.
Zittoun, P., Demongeot, B. (2010) ‘Debates in French policy studies: From
cognitive to discursive approaches’, Critical policy studies 3(3–4), 391–406.
Zittoun, P., Peters, B.G. (2016) Contemporary Approaches to Public Policy: Theories,
Controversies and Perspectives, Basingstoke: Palgrave Macmillan.
28
Part One
The styles and methods of
public policy analysis
29
TWO
On the path to public policy analysis:

an ‘administrative science’ between reform
and academy
Renaud Payre and Gilles Pollet
The simple question to which this first ‘sociohistorical’ chapter wishes to respond is
the following: what types of studies, writings and reflections existed on the theme
of public administrations and their policies even before the institutionalisation of
public policy analysis per se in France? And how did this corpus of understandings
and questioning, of knowledge and know-how, influence the constitution of
this academic speciality, or indeed not? In other terms, in this chapter we would
like to examine the sociogenesis of an academic discipline by illustrating the
filiations, borrowings, hybridisations and co-constructions which contributed to
the public policy analysis1 ‘à la française’ of current times. Our hypothesis is that,
in the French case, public policy analysis, primarily inspired by North America,
also developed in response to the relative failure of ‘administrative science’ to be
institutionalised, and from the science’s foundations and eventual prolongations.
Indeed, in France, public policy analysis took over from an ‘administrative
science’ which tried to develop over almost a century but which, unlike in the
United States, was never fully institutionalised and professionalised. This occurred
for two main reasons. First, due to the weight of law in the training of French
elites, this administrative science constantly tended to fall under the specialty of the
administrative law taught in the Facultés.2 It was also the result of the fundamental
role played by some senior civil servants, jurists by training and legal practitioners
– notably state councillors – in the determination of jurisprudence and a form
of litigation science, as well as reflections and policies pertaining to the reform,
modernisation and operation of the state and public administrations. These senior
civil servants considered themselves to be the only legitimate actors operating in
this domain, although they were also state actors and could thus be referred to
as both ‘judge and jury’ of politico-administrative action. It is necessary to make
a specific inventory of what this ‘administrative science’ bequeathed to public
policy analysis in France, as this historical episode constitutes an essential element
to understanding the sociogenesis of French public policy analysis.
At the origin of the public policy analysis which would develop in France
from the 1970s and 1980s onwards, particularly within political science as an
intersection for academic disciplines, we indeed find, in addition to and next to
a great number of Anglo-Saxon and especially North-American contributions
(Mény and Thoenig, 1989), an entire series of empirical and theoretical research
31
and forms of problematisations linked to the (effective) operation of public

administrations (developed in Michel Crozier’s Centre for the Sociology of
Organisations (Centre de Sociologie des Organisations [CSO]). This initial observation
leads us back to the history of a science of public policy even before its academic
institutionalisation. The approach which aims to reconstitute a non-disciplinary
history of these disciplines is already well-known (Burrow et al, 1983; Ross, 1991).
These works instead impel us to propose a non-retrospective socio-history of
the knowledge enterprises of politico-administrative action and their link to the
activities of administrative and political practitioners. From this angle, attention
can be drawn to fragile scientific projects, diverse intellectual enterprises and
their competitive effects, distinguishing the consolidation of certain of them
and the possible eviction of others. Starting from this basis, we will equally take
into account the academic disciplines, as well as the projects or even the simple
claims, of an administrative science.
Our intention is not to reread the past in the light of current times, nor to control
the memory of the discipline in order to decide the future or to finally determine
some effects of its institutionalisation. Such a history was already produced just
after the Second World War (Langrod, 1966; Prélot, 1961) and sought a genesis
for the discipline by evoking the ‘science of the police’ from the seventeenth
and eighteenth centuries, developed by German Kameralists and by several great
French forerunners – like Nicolas de La Mare in his work Police Treaty (Traité de
la Police). But, above all, this particular account dwelt on the nineteenth century
in order to put forward a real pantheon of founding fathers of this science: for
example, Tocqueville in France, von Stein in Germany, Wilson in America. Its
chief aim was to convince us that the discipline truly existed. The approach which
we would like to establish is different. It instead proposes a new reading of the
collection of documents and discourses related to the administration, its reform
and its possible rationalisation and modernisation, through the tenuous assertion of
a genuine ‘administrative science’ which, in fact, would never be truly legitimised
and institutionalised – at least not before the 1940s. This approach involves an
examination of the different settings of text production and the configurations
of actors and institutions which then backed and diffused these texts. Ultimately,
‘administrative science’ will be addressed here as a vast reformative movement
which attempted to institutionalise itself, in particular at an academic level, and
within which the French actors were connected to a group of networks which
appeared, from the twentieth century onwards, to be largely internationalised,
even transnational, and, more specifically, transatlantic.
For more than a century, ‘administrative science’ constructed itself as a national
and transnational cause with, in the French case, the domination of jurists and state
Councillors limiting its institutionalisation as a practical science. At an academic
level, the turning point of the 1940s, and most importantly post-Second World
War, was marked by a new context, more favourable to an administrative science
which was seen as a politico-administrative reform science, but which would still
only have rare openings into the academic and scientific sphere. This incomplete
32
On the path to public policy analysis
academic institutionalisation, particularly through the critical debates which

it provoked, could appear as the premises and the ground for both the future
development of a new specialty, largely inspired by North-American writers, and
the complex debates between supporters of a science of government and Public
Administration studies: public policy analysis.
‘Administrative science’ as a national and transnational cause

(1830–1930)
In many ways, ‘administrative science’ constitutes one of the essential matrixes
of modern public policy analysis. If one wishes to grasp the complexity of its
development, it is necessary to think outside uniquely academic spheres and
to seriously consider the collection of discourse and actions related to the
administration and its possible rationalisation. In France, from the 1830s, a
milieu dedicated to the production of knowledge specific to the administration
appeared. Its institutionalisation, throughout the nineteenth century, occurred
via the publication of reviews and dictionaries, and through the organisation of
teachings, training schools and the activities of diverse learned societies. It also
progressed, at the beginning of the twentieth century, by virtue of the circulation
of numerous actors and ideas linked to the construction of a transnational space of
international congresses, in which the weight of American reformers – disciples
of a science of public administration – would become increasingly important.
This milieu of reform was initially structured around editorial spaces. Thus, in
the 1830s, the first reviews pertaining to administrative duties were published.
This was the case of the Legislation and Jurisprudence Review (Revue de législation
et de Jurisprudence), founded in 1834 by the jurist, economist and politician
Louis Wolowski (1810–76). This review clearly supported the preparatory
education projects for the Civil Service. In 1839, the Administration Review
(Revue administrative) was also created. It would be carried on in 1846 by law
professor and State Councillor Alfred Blanche (1816-1893) and the publisher
Paul Dupont (1796-1879), until 1849, when these two authors published the
General Administration Dictionary (Dictionnaire général d’administration). Three
decades later, in 1878, the statistician and economist Maurice Block (1816-1901)
founded the General Administration Review (Revue générale d’administration),
which was published until 1929 and marked the establishment of the enterprise
of this administrative milieu (Payre, 2013). After a year in operation, even before
the regime became a republic, the review was entrusted to the civil servants
from the Ministry of the Interior’s very first departmental division office. The
mission assigned to this review was thus: ‘to educate, to control, to critique, to
indicate progress to be made, reforms to be accomplished’. The reviews often
accompanied other publications of breadth, such as dictionaries. In addition to
the dictionary edited by Blanche and Dupont in 1849, we cite in particular that
of Block, published in 1856. The objective of Blanche’s dictionary was clearly
expounded in its manifesto: ‘to offer administrators special, specific, and technical
33
notions, as they have the right to demand them; to speak without dryness to the
people of the world of things which are far from being incompatible with a form
of charm and interest, to tirelessly initiate them into a new science for them and
to inspire their desire to explore it further’.
Projects for a school of administration also constituted one of the major tangible
objectives of this milieu. The history of a national school of administration
is one of aborted projects (from Condorcet to Victor Duruy, via Portalis).
Nevertheless, under the July Monarchy (1830–48), the project was on the verge
of materialising. In 1840, under the auspices of Public Education Minister Victor
Cousin (1792–1867), a study was conducted on administrative teachings in
Germany. In 1843, the jurist Edouard Laboulaye (1811–83) published a summary
of the study in the Revue de législation et de Jurisprudence and the report was then
carried on by Hippolyte Carnot, Public Education Minister (1848). However,
the project concerning the creation of a special school of administration had not
yet been presented to the Chambers when the Revolution broke out in 1848.
All the same, we can point to the existence of diverse teachings, for example at
the Egyptian school located in Paris (École egyptienne de Paris), at which the
State Councillor and politician Sébastien Joseph Boulatignier (1805–95) had
already created administrative law classes. Several years later, the State Councillor
Léon Aucoc (1828–1910) would make these classes the matrix of the school of
administration. In fact, in 1848, the government decreed that ‘an administrative
school destined to recruit for the diverse administration branches, until present
devoid of preparatory schools, will be established along the same basis as that
of the École polytechnique [Polytechnic School]’. However, this school would
only host two student cohorts and would close the following year. Nonetheless, a
proportion of the students would seek to obtain the reopening of a similar training
institution. In this respect, until 1848, the Association of former students of the
school (Association des anciens élèves de l’École), would play an extremely important
role. On the other hand, having opened its doors in 1872, the Independent
School of Political Sciences (École Libre des Sciences Politiques) – referred to by
Emile Boutmy as the State Sciences University (Université des sciences d’État) –
was, for a long time, regarded as a competitor for the eventual national school
of administration. In 1877, Hippolyte Carnot submitted a new proposition for
the creation of a national school of administration to the Senate, even though
the Independent School of Political Sciences had opened five years previously.
The failure of this proposition marked the rapprochement between the financial
support for a national school of administration and the Saint Guillaume Street
School (École de la rue Saint Guillaume).3
In conjunction with the reviews, dictionaries and schools, the learned societies
(the Political Economy Society (Société d’économie politique), the Statistics Society of
Paris (Société de statistique de Paris), the Society of Comparative Legislation (Société
de législation comparée), the Society for the Encouragement of National Industry
(Société d’encouragement de l’Industrie nationale), among others) also contributed
to the production of administrative knowledge. The Société de statistique de Paris
34
(closely linked to the Economists’ Journal [Journal des Économistes]) was of particular
importance. At its very origin, it was the extension of the network of liberal
economists, brought together since 1842 in the Société d’économie politique. At the
request of Michel Chevallier (1806–79), a graduate of the École polytechnique,
a mining engineer, and also the Chairman of Political Economy at the French
College University (Collège de France), the main members of the Société de statistique
came to constitute the core of the new society. Among the most active, some had
directly participated in the French General Statistics Society (Statistique Générale
de France), notably Alfred Legoyt (1812–85) and Maurice Block. Yet, looking
beyond these learned societies, the Academy of Moral and Political Sciences
(Académie des Sciences Morales et Politiques) was also an institution producing
administrative knowledge, especially its political economy division – open to
current or former civil servants. In connection with these societies, the science
of administration – not yet named ‘administrative science’ – was produced in
international congresses, first and foremost featuring congresses on statistics, in
which the actors producing administrative knowledge were invested, and from
which they developed an understanding of foreign works.
It was precisely this transnational space which transformed itself in the early years
of the twentieth century. One congress in particular draws our attention: the first
international conference for administrative sciences, which was held in Brussels
in 1910 and was aimed at both ‘savants’ and ‘men of action’. The objective of
the congress was to discuss the forms and methods of government – central and
local – as well as the relationships established between the different administrative
authorities. In Brussels, the Belgian President and politician Georges Cooreman
(1852–1926) already sought to define the subject matter of the congress.
According to him, ‘administrative science’ would first espouse an understanding
of law, but it would not be limited to this knowledge and should also constitute
a ‘technical science connected to different branches of the administration:
financial, economic, social, health, commercial, design, etc.’ Furthermore, it
would be a ‘government science’, developing ‘an understanding of social facts,
a perfect knowledge of the mentality of individuals, a fair appreciation of their
conception of discipline and order, a faithful assessment of their powers and also
of their taxable resources’. Therefore, above all, ‘administrative science’ would be a
practical science oriented towards the improvement of administrative methods and
composed of a range of pre-existing knowledge. But the Permanent Commission
for International Administrative Sciences Congresses (Commission permanente des
Congrès internationaux des sciences administratives) was interrupted during the First
World War and the first post-war congress took place in Brussels in 1923, then
in Paris in 1927. They were still debating over municipal administration, in the
fourth international congress in Madrid in 1930. It was during this last congress
that the (Commission permanente transformed itself into the International Institute
of Administrative Sciences (Institut international des sciences administratives) (Payre,
2011).
35
Who were the French delegates at these congresses? We can essentially identify
some of the State Councillors and the law professors, in particular from the Law
Faculty (Faculté de droit de Paris).4 To begin with, Henry Berthélemy (1857–1943),
the then Dean of the Law Faculty, was also the Vice-President of the Commission
permanente until 1930. He was replaced by Joseph Barthélemy (1874–1945), who
had tenure as a professor at the Law Faculty from 1918. Furthermore, during
the international congress in Paris in 1927, the organisation of the event was
partially ensured by the Law Faculty. During the 1930s, the position of Secretary
for the French branch of the International Institute returned to an ‘activist’ for
administrative science and State Councillor: Henry Puget (1894–1966). After
the Second World War, Puget joined the Board of Directors of the International
Institute and presided over the commission of scientific works. Even without
listing other politicians and town planning specialists, this portrait of the French
delegates from the International Institute of Administrative Sciences indicates
the porosity between administrative law and the young administrative science,
which clearly appeared to be a French, and partially European, specialty. The
frontier with legal science, still poorly defined in the interwar period, unveiled
an administrative science of which the practical objectives still often amounted
solely to administrative law.
During the 1930s, the architects of this administrative science, conceived
in essentially European congresses, came into contact with the American
reformers, the original professionals of a science of public administration which
had developed and institutionalised at the same time on the other side of the
Atlantic. This ‘science’ of public administration, which had staked its claim in
the United States at the end of the nineteenth century, had not solely developed
in the universities. A large progressive reform movement also contributed to its
formation, in particular though the founding of municipal research offices (the
first in New York in 1907), then an Institute of Public Administration (1918),
a governmental research association (1924) and even a Public Administration
Clearing House, created in 1932 in Chicago and which, seven years later,
would house the American Society of Public Administration. In contrast, the
American delegates of international congresses then discovered, in the middle
of the 1930s, a European administrative science, fundamentally created by
administrative law professors and very far from the practical aims of its homologue
across the Atlantic. They were, moreover, struck by the limited influence of the
International Institute. Voyages, and more informal exchanges outside meetings,
contributed to persuade the European officials of the pertinence of the Americans’
criticisms. It was necessary, in fact, to professionalise the international meetings
of administrative sciences and to unify the work methods. It was this encounter
with the American science of public administration – largely developed outside
the legal sphere and with an even stronger anti-legal orientation – which then
created a frontier between administrative law and administrative science, which
had been unstable up until this point.
36
The Americans’ participation in the Institut grew throughout the 1930s.

Criticising the weight of administrative law in the congresses’ work, the Americans
tended to diminish its importance in favour of questions related to the concrete
administration: budgetary practices, the rationalisation of structures and methods,
questions related to organisation and decision-making, and so on. The academic
Rowland Egger (1899–1980), future co-author of the famous work The President
and Congress (1963), only needed to participate in the regional congress of
administrative sciences, held in Brussels in June 1935, to understand the use of
the term ‘administrative sciences’ in continental Europe and, in particular, France.
The work sessions in which he participated persuaded him that the Institute, in
his eyes, was too preoccupied with minor legal problems, administrative theory,
even documentary issues. The research was not sufficiently oriented towards
administrative practice. As the task of the American architects of the science
of public administration was, above all, to conduct studies for the purposes of
practical research, Egger was convinced of the necessity to strengthen the British
and American presence within the Institute’s Board of Directors. The American
equivalents, members of the Public Administration Clearing House, also wished
to place new men at the head of common services such as the International
Union of Cities (Union internationale des villes) and the International Institute of
Administrative Sciences. They invited several young British, Belgian and Dutch
academics to Chicago so that they could observe the work being accomplished
in the Public Administration Clearing House and take inspiration from tried and
tested methods from the other side of the Atlantic.
For more than four decades, in the United States, the science of public
administration was the subject of a both political and scientific quest which, above
all, sought to make administrative work effective. This concept of effectiveness,
like the notion of separation of the administration and politics (Wilson, 1887;
Goodnow, 1900), was at the very basis of administrative theory. The act of
administration appeared to be positive and free from value judgements. As for
the administration itself, it had to be rational in order to be a subject of science
and to enable the legitimisation of ‘administrative science’, called for by all
but with different conceptions and in very different contexts. Nevertheless,
this science of public administration was part of a larger reflection pertaining
to general administration issues, according to the French mining engineer
Henri Fayol (1841–1925). His 1916 work, General and Industrial Administration
(Administration industrielle et générale), somewhat forgotten in France, was reread
and commentated during the 1930s and 1940s by eminent Anglo-Saxon public
administration specialists. In this respect, we must cite the Briton and business
consultant Lyndall Urwick (1891–1983), who directed the Management Institute
(Institut de management) in Geneva from 1928, and also the American ‘senior civil
servant’, Luther Gullick (1892–1993), an advisor to President Roosevelt and
President of the Institute of Public Administration, with whom Urwick would
create the well-known Administrative Science Quarterly in 1956.
37
Paradoxically, it is therefore from the United States, and Anglo-Saxon social

scientists and practitioners, that Fayolism and studies into general administration,
in particular public administration, were regenerated and diffused during the
interwar period. We then saw the establishment, in transnational spaces, of a battle
of influence between social and political scientists and Anglo-Saxon reformers
– especially those of American nationality – on one side, and European jurists
and senior civil servants – largely of French nationality – on the other. The
European actors, dominant within their national scene and convinced of their
legal legitimacy, had to deal with the practitioners and analysts of administration
and public policy who were in the process, with other researchers and actors –
of whom some were very critical in this regard – of creating the basis of policy
analysis, including the dimension of its academic discipline.
The uncompleted institutionalisation of ‘administrative science’

and the foundations of public policy analysis (1930s and 1940s)
In the French case, ‘administrative science’, which involved a science of, and
for, public policy, experienced a particularly academic, stymied, and eventually,
for the most part, uncompleted, institutionalisation which would, all the same,
constitute one of the grounds on which public policy analysis ‘à la française’
would develop, on the fringes of the universities.
In France, ‘administrative science’, especially its academic dimension, would
indeed be influenced and partially contested by the diverse reform projects
of the state and the Civil Service, which emerged in the interwar period and
lasted until Maréchal Pétain’s Vichy regime during the German occupation
(1940–44). The willingness for rationalism and modernism was thus particularly
manifested in a constellation of ‘elite technicians’, which we can divide into two
main groups of actors. On one hand, there were the private sector engineers
and employers, as well as state engineers, who placed the economic dimension,
as well as quantifiable, rational and, according to them, truly ‘scientific’ data, at
the heart of their reflections. Tainted by Fayolism, economic and administrative
expertise could only satisfy a form of ‘administrative science’ which, according to
them, would neglect the private sector and could be associated with a politico-
administrative system founded on the political parties and a parliamentarianism
which they frequently denounced. This may have led some of these actors to
advocate for technical reforms, often anti-liberal and sometimes even authoritative,
which may have prefigured the Vichy regime, illustrated by certain propositions
of ‘French Re-establishment’ proposals (‘Redressement français’), created in 1925,
and even the group X-Crise in the 1930s. On the other hand, for the state’s elite
administrative corps, notably some members of the State Council and public law
academics, it was conceivable to imagine a technical reform of the state, remaining
for the most part compatible with the values of the liberal and representative
democracy. In this context, as an academic specialty and practical know-how,
‘administrative science’ represented a path to a possible institutional and technical
38
reform of the state, retaining the form of a parliamentary regime but reshaping
it (Brun, 1985; Pollet, 1999).
In French universities, however, in the last decades of the nineteenth century,
administrative law imposed itself as the primary science of public administration. It
had notably benefited from the dynamic triggered by the creation of a competitive
Civil Service exam specific to public law in 1895 and by the concomitant
development of constitutional and administrative law teachings in the Facultés.
Nevertheless, as we have seen, an administrative science, in addition to its specific
knowledge and know-how, had been developed in other spheres. Civil servants,
practitioners, ‘technicians’ and senior civil servants took the initiative to develop
– often outside universities, in specialised periodicals, congresses or associations –
an ‘administrative science’ based on different disciplines. We have also examined
the very extent to which the practices of a North-American science of Public
Administration played a leading role, as a unifying force or counter-model, in all
of these practical experiments and intellectual exchanges. It was through these
men and these examples that the idea of a practical science of administration,
often described as a technical science, was established. In France, the theoretical,
then practical, construction of the legal State (Carré de Malberg, 1920) further
reinforced the place of both public and administrative law in the hierarchy of
legal disciplines (Chevallier, 1992). At the end of the 1930s, it seemed recognised
that ‘administrative science had taken on a very technical character and the
administrative career demanded…numerous and important legal, economic or
social knowledges’. These remarks can be found in the explanatory statement
for the draft bill which the French government submitted to the Chamber of
Deputies on 1 August 1936; a draft bill authorising the creation of a national
school of administration. ‘Administrative science’, seeking scientific legitimacy,
wished to rationalise its subject matter.
Nonetheless, very quickly, the division between administration and politics was
judged obsolete by certain researchers specialised in administrative studies. This
tended to be affirmed in the 1940s, particularly after the Second World War,
against the backdrop of the general reform of the state, the Civil Service and
the training of elites, in which ‘political and administrative sciences’ would play
an essential role (Langrod, 1956–57). In the dynamic interplay unleashed by the
numerous debates and propositions during the interwar period, several reform
projects for the state and the Civil Service were developed by the Vichy regime,
by the National Council for Resistance (Conseil National de la Résistance), and even
by the Grands Corps5 themselves, notably the State Council (Baruch, 1997). This
saw the opposition, and coexistence, of technical and technocratic conceptions
of public policy, modernist and democratic desires, and elitist and corporative
notions. Although proposed solutions could be radically different, all those
involved seemed to agree on the triple assessment of the ill-adapted training of
civil servants, the undemocratic recruitment system and the suboptimal operation
of the politico-administrative apparatus. In the context of the Liberation of Paris
and the reconstruction of the nation, ‘administrative science’, often encompassed
39
by the notion of ‘administrative and political sciences’ or even ‘political sciences’,

finally believed that it had found a favourable ground for development.
Directly after the Second World War, the academic institutionalisation of the
science of administration was effectively taking place. In France, as it was at an
international level, government sciences became the prerogative of the higher
education and research sector, integrated into universities and, in the French
example, more directly into the Grandes Écoles,6 the research organisations (the
National Centre for Scientific Research (Centre National de la Recherche Scientifique
[CNRS]) was created by decree in 1939), and even into emerging international
academic associations. In 1945, the teaching of ‘political sciences’ was diffused
throughout the French territory. The government directly participated in the
implementation of the teaching of administrative and political subjects. At the
end of the 1940s, the International Political Science Association (AISP) came
into existence in conjunction with the United Nations Educational, Scientific
and Cultural Organization (UNESCO) – the international organisation which
had previously conducted a study into research, the method and teaching of
subjects relative to politics. In newly liberated France, the reform of the state,
the selection of senior civil servants and the restructuring of the higher education
system, tended to consolidate ‘administrative science’ at an academic level.
The entire training programme for elites was then reconsidered in order to break
with a system which would have led to defeat through its collaboration with the
Vichy regime. Characterised by a certain cultural and historical defiance vis-à-vis
the universities, and approving the typically French model of the Grandes Écoles,
in 1945 the provisional government, presided over by General de Gaulle, created
the National School of Administration (École nationale d’administration [ENA])
by law. This school was intended to form the new, national and multidisciplinary
administrative elites, and it was hoped above all that it would be independent of
territorial and sectorial issues. The aim was to break with forms of populism,
localism and corporatism, which were considered to have represented the negative
side of the exiting Third Republic. Through this general reform to the training of
the Nation’s elites, in particular the state’s Senior Civil Service, the restructuring
of ‘political science’ teaching and research enabled administrative science to find
its own space and a definite, although fragile, academic legitimacy. It was within
the Grandes Écoles rather than at the universities that this scientific, technical and
managerial aggiornamento would occur. The École Libre des sciences politiques
was thus dissolved and became, in 1945, a public establishment of higher learning
and research, under the denomination of the Institutes of Political Studies (Institut
d’Études Politiques [IEP]) of the university of Paris. A National Foundation for
Political Sciences (Fondation nationale des sciences politiques [FNSP]) of private law
was also put in place with the goal to ‘foster the progress and diffusion, in France,
in the Empire and abroad, of political, economic and social sciences’ (Article 1
of ruling 45-2284, 9 October 1945). The FNSP presided over the creation of
diverse ‘administrative science’ research centres in Paris, then regionally, once
again at the intellectual and physical margins of the universities.
40
The Administrative Research Centre (Centre de Recherches Administratives

[CRA]) was thus developed from 1946, directed by the State Councillor and Paris
IEP professor of law, Henry Puget. The latter tried to establish a link between
the diverse institutions which came into existence during the interwar period:
the Centre for Administrative Research (Institut de droit compare) from the Law
Faculty, the Institute for Comparative Law (Institut des techniques de l’administration
publique) (Bergeron, 1969; Weexsteen, 1999), as well as foreign and international
institutions such as the International Institute of Administrative Sciences and its
national branch, the French Institute of Administrative Sciences (Institut Français
des Sciences Administratives), of which he would be the General Secretary for
more than 30 years. The CRA was oriented both ‘towards the development of
scientific knowledge and towards the improvement of methods and practices’
(Puget, 1950). Luc Rouban revealed that the CRA had obtained the heritage
of the Centre for the Study of Administrative Issues (Centre d’étude des problèmes
administratifs), housed by the Alex Carrel Foundation (Fondation Alexis Carrel)
during the Vichy regime (Rouban, 2005). At the end of November 1946, a
development committee was created to ‘orient the CRA’s work, including René
Cassin, Vice-President of the State Council, André Siegfried, President of the
FNSP, the Director of the ENA, the Director of the IEP, the Director of the
Civil Service, the Director of the Budget, the Secretary-General of the National
Federation of Civil Servants (Fédération nationale des fonctionnaires) (the Leap), the
President of the Union of Cities and Local Powers (Union des villes et pouvoirs locaux)
(Marrane), Marcel Waline and Baumgartner, President of the Crédit national.7
A new stage in administrative research has thus begun’ (Rouban, 2005, 141).
In September 1947, the Nation’s official journal also made public the creation
of the Institute of Public Administration Techniques (Institut des Techniques de
l’Administration Publique [ITAP]), of which the project dated back to 1935, the
year during which the National Council for French Organisation (Conseil national
de l’organisation française), chaired by Henry Puget, decided to found a permanent
division of public administrations. Furthermore, new specialised periodicals would
also come into being, such as the Revue administrative, founded in 1948.
That same year (1948), after Paris and Strasbourg three years earlier, additional
Instituts d’Études Politiques were created in the provinces: Bordeaux, Grenoble,
Lyon and Toulouse, then Aix-en-Province in 1956. This last Institut was preceded,
in 1951, by the establishment of a centre for political and administrative studies,
at the instigation of the Professor of Public Law Paul de Geouffre de La Pradelle
(1902–93). It was even envisaged, at one time, to name the Instituts d’études
politiques ‘administrative science institutes’. The ruling of 1945 also mentioned, for
that matter, that the ‘institutes have the mission to provide students, whether or
not they are designed for the Civil Service, with general political or administrative
culture. They will do this in a spirit of independence and impartiality which are
characteristic of the universities’. Indeed, in all of the IEPs, which opted for a
multidisciplinary cursus in human and social sciences, from the very beginning
we find teachings most often grouped under the term ‘administrative and political
41
sciences’, composed of classes on political and administrative institutions, on

public and administrative law, on the Civil Service, and all from a French, but
also comparative, perspective. For over 20 years, Henry Puget also gave a highly-
reputed class at the Paris IEP, on foreign administrative and political institutions.
On the university side, it was necessary to wait until 1954, and the reform of legal
studies, for political science classes to be introduced into law faculties, among
which we find several teachings on ‘administrative science’ (Sadran, no date, 131).
The specialty, however, remained poorly defined and only slightly
institutionalised at an academic level. In the middle of the 1950s, in an article
entitled ‘Administrative Science or Administrative Sciences’ (‘Science administrative
ou sciences administratives’), evoking a ‘administrative science’ more consolidated
than that of the interwar period, Georges Langrod, a former lecturer in public
administration and administrative law at the University of Warsaw and a
researcher at the CNRS (from 1948), defined the discipline as follows: ‘The
subject of administrative science covers administrative facts, that is to say, as much
institutionalised adjustments as natural social phenomena, in direct relation to
the existence and the operations of public administrations’ (Langrod, 1956–57).
Furthermore, he wished to respond to the following question: ‘The public
administration (its organisation and its activity), is it (or could it legitimately be)
the subject of a science or only of an art?’ (Langrod, 1956–57). He thus defined
administrative sciences as both a science and art, and as a group formed from all
of the disciplines which study problems relative to administrative phenomena.
In the work Mélanges Georges Langrod, Henry Bergeron, then an administrative
practitioner and a civil servant for the city of Paris, again clarified the conception
of ‘administrative science’ of his former professor. Starting with the notion of
the administrative fact, it must, by explaining the data, ‘determine the routes of
action, pre-empt the practice and render the results usable’ (Boulet, 1980, 77).
In France in the 1940s and 1950s, therefore, a homology existed between the
consolidation of ‘administrative science’ and the reputation of its subject matter.
In fact, as ‘administrative science’ was becoming academically institutionalised,
it also sought to legitimise concrete administrative policy. Those advocating for
the recognition of this type of discipline also wished to convince others of the
transformation of the public power’s policy and, above all, of the growing role
that the administration had assumed in the management of people and matters.
Imposing the notion of the existence of a science of administration supposed
that the administration itself had become ‘a great technique, a science, or more
precisely, a group of sciences’. Therefore, ‘administrative science’ claimed to be
a group of techniques, knowledge and know-how, made available to elected
officials as well as civil servants. It was not uniquely defined as a ‘science’, as a
simple academic and university specialty, but also as an ‘art’ – understood to be
a collection of know-how and techniques destined for civil servants. Here, the
two dimensions – academic and applied, normative and prescriptive – were still
strongly associated. Therefore, one of the principal objectives of these works and
research, pragmatic or rather academic, was to assert itself by specialising and
42
improving its subject matter. Nevertheless, this great aspiration to scientificism

quickly became aware of its limits and ‘administrative science’ would never manage
to exist as an autonomous specialty or discipline, nor be institutionalised from a
corpus of understanding and from a system of consistent and well-known actors at
an academic level. The double constraint of the law (especially administrative) as
a dominant academic discipline and professional practice – as a ‘corps of rule and
corps of professionals’, to employ Jacques Chevallier’s expression (Chevallier, 1993)
– and the weight of the jurists and State Councillors as practitioners and producers
of the same law, through jurisprudence and in particular the Council’s rulings,
would finish by slowing, then making this ambitious project almost disappear.
It would be carried on and reconfigured from the 1960s in another context and
with other aspirations, in particular playing the role of an interface between law
and social sciences (Chevallier, 1993).
The evolutions of ‘administrative science’ in the French academic context
are also illustrated by the awareness of international, and especially American,
influences, as we have already noted. Indeed, in the United States, from the
1940s, a debate began on the scientism of a discipline centred on the study of
public administration. This debate, essentially university-related, also concerned
Europe, and a fortiori, France, as there was a greater number of international
exchanges on the subject of political sciences after the Second World War. In
1948, during the first international meeting on the subject of political science at
UNESCO, Charles Merriam evoked the ‘considerable progress’ demonstrated
in the United States following ‘the abandonment of the study of forms and
institutions in favour of organic processes of different behavior types’ (Merriam,
1950). The scientific character of this knowledge of administration was one of
the main leading arguments in the indictment that advocates of behaviorism
addressed at the realist approach.
In the United States, as on the international scale, the 1940s marked the arrival
of a historical moment in which it seemed possible to construct a normative
‘administrative science’ – in the sense of a science of public administration – largely
situated outside the academic system (Lindblom, 1997). Therefore, Raymond
Seidelman studied the interweaving of the reform projects and the first political
science works in the United States at the turning point of the century. Through
that which he names the ‘disenchantment of realists’, starting with the trajectories
and works of certain academics such as Charles Merriam, he identified that at the
end of the Progressive Era (corresponding with the end of the First World War),
following the failure of reform enterprises, the ‘political scientists’ withdrew to
the universities and sought to strengthen the ‘scientific’ foundations of political
science. Their ‘scientism’ resided in the search for normative and prescriptive
principles which aimed to explain and to transform the attitudes of citizens.
Seidelman pondered over the disappearance of this type of conception of political
science at the time of the Second World War. The behaviorists of the 1950s
would then be the first to rank Merriam among the ‘primitive scientists’, or even
the ‘utopian dreamers’ (Seidelman, 1985). As a reference, this ‘potential science’
43
adopted a clear distinction between administration and politics. The politics would
be focused on opinion, while the administrative spirit would be characterised by
both reason and technical aptitude. This supposed apoliticality was nonetheless
markedly discussed in the United States in the 1940s. The science of public
administration was brought into question by young arrivals to the academic scene,
after having barely submitted their PhDs. The arguments in favour of a greater
positivist repositioning then ultimately advocated professionalising the science
of public administration at an academic level, to create an academic discipline
out of this government science; a practical science of which the works were
above all accomplished by engaged practitioners or reformers, as we have seen.
To some extent, the United States succeeded in creating, during the interwar
period, a science of public administration within the universities, whereas the
French failed to institutionalise their ‘administrative science’, which ‘was reduced
to the administrative law with which it was confused and in which it retired’
(Chevallier,1993, 16).
The development and the academic institutionalisation, within French political
science, of public policy analysis during the 1970s and 1980s would therefore
mark both the completion and the closure of an ‘administrative science’ which
never really succeeded in establishing itself at an academic level or in the politico-
administrative universe. Despite its unfulfilled institutionalisation, this ‘discipline’
nevertheless provided a number of intellectual frameworks, programmatic research,
and concrete policies, pertaining to the government and modern public policy. By
attempting to partially free itself from administrative law, from the constraints of
the purely legal approach and the jurisprudence practices of the State Councillors,
all while facilitating fruitful transnational and in particular transatlantic exchanges,
this unrealised ‘administrative science’ constituted both a type of counter-model
and also certain foundations of public policy analysis ‘à la française’, of which a
specific inventory remains to be detailed.
Notes
1 Public policy analysis here refers both to its academic research dimension (policy process studies)
and its specialist dimension oriented towards concrete public policy (policy analysis).
2 Prior to 1968, French universities were referred to as Facultés.
3 The Saint Guillaume Street School was another name for the École Libre des Sciences Politiques.
4 In the University of Paris (Université de Paris).
5 Elite state bodies.
6 Elite French higher education establishments.
7 The Crédit National is France’s main national bank.
44
References
Baruch, M.-O. (1997) Servir l’Etat français: l’administration en France de 1940 à
1944, Paris: Fayard.
Bergeron, G. (1969) Fonctionnement de l’État, Paris: Librairie Armand Colin.
Blanche, A. (1849) Dictionnaire général d’administration, contenant la définition de tous
les mots de la langue administrative, Paris: P Dupont.
Block, M. (1856) Dictionnaire de l’administration française, Paris: Berger Levrault.
Boulet, L. (ed) (1980) Science et action administratives, Paris: Editions d’Organisation.
Brun, G. (1985) Technocrates et technocratie en France (1914–1945), Paris: Albatros.
Burrow, J., Collini, S., Winch, D. (1983) That Noble Science of Politics: A Study in
Nineteenth-century Intellectual History, Cambridge: Cambridge University Press.
Cahiers de la FNSP (1965a) ‘Administration traditionnelle et planification
régionale’, 135, Paris: Armand Colin.
Cahiers de la FNSP (1965b) ‘La planification comme processus de décision:
Compte-rendu des travaux du Colloque de Grenoble’, 140, Paris: Armand
Colin.
Carré de Malberg, R. (1920) Contribution à la théorie générale de l’Etat, Paris: Sirey.
Chevallier, J. (1992) L’Etat de droit, Paris: Montchrestien.
Chevallier, J. (1993) ‘Le droit administratif entre science administrative et droit
constitutionnel’, Le droit administratif en mutation, pp. 11–40, Paris: Presses
Chevallier, J., Lochak, D. (1980) La science administrative, Paris: Collection ‘Que
sais-je?’, Paris: Presses universitaires de France.
Dahl, R.A. (1947) ‘The science of public administration: Three problems’, Public
Administration Review 7(1) (Winter), 1-11.
Goodnow, F.J. (1900) Politics and Administration: A Study in Government, New
York: Macmillan.
Grunberg, G. (2004) ‘La recherche à Sciences Po’, La revue pour l’histoire du
CNRS, 11, http://histoire-cnrs.revues.org/680
La Mare, N. de (1705-1738) Police Treaty (Traité de la police), Paris: J. et P. Cot,
M. Brunet, J.-F. Hérissant (eds).
Langrod, G. (1956–57) ‘Science administrative ou sciences administratives?’,
Annales Universitatis Saraviensis 6, 92–125.
Langrod, G. (1966) ‘La science administrative et sa place parmi les sciences
voisines’, Traité de science administrative, pp. 92–123, Paris: Mouton.
Lindblom, C.E. (1997) ‘Political Science in the 1940s and 1950s’, Daedelus: Journal
of the American Academy of Arts and Sciences 126(1) (Winter), 225–52.
Mény, Y., Thoenig, J.-C. (1989) Politiques publiques, Paris: Presses universitaires
de France.
Merriam, C. (1950) ‘La science politique aux États-Unis’, UNESCO, La science
politique contemporaine: Contribution à la recherche, la méthode et l’enseignement,
Paris: UNESCO.
45
Payre, R. (2011) ‘L’espace des circulations. Réseaux d’acteurs et fabrique

transnationale des sciences administratives (années 1910–années 1950)’, in
F. Audren, P. Laborier, P. Napoli, J. Vogel (eds) Les sciences camérales: activités
pratiques et dispositifs publics, pp. 283–306, Paris: Presses universitaires de France.
Payre, R. (2013) ‘La compétence contre la politique: Milieu administratif et
élaboration des savoirs légitimes de l’administration dans la France du second
XIXe siècle’, in Y. Deloye, O. Ihl, A. Joignant (eds) Gouverner par la science:
perspectives comparées, pp. 25–40, Fontaine: Presses universitaires de Grenoble.
Pollet, G. (1999) ‘Technocratie et démocratie. Elites dirigeantes et réforme
technicienne de l’Etat dans la France de l’entre-deux-guerres’, V. Dubois, D.
Dulong (eds) La question technocratique: de l’invention d’une figure aux transformations
de l’action publique, pp. 35–53, Strasbourg: Presses universitaires de Strasbourg.
Prélot, M. (1961) La science politique, Paris: Collection ‘Que sais-je?’, Presses
Puget, H. (1950) ‘Le Centre de Recherches Administratives’, La Revue
Administrative 13 (January–February), 107–10.
Ross, D. (1991) The Origins of American Social Science, Cambridge: Cambridge
University Press.
Rouban, L. (2005) ‘Réformer l’Etat: Henry Puget et la Fondation Alexis
Carrel 1943–1946’, La Revue Administrative 344 (March), 127–42.
Sadran, P. (no date) Public Administration Programmes in France, http://unpan1.
un.org/intradoc/groups/public/documents/NISPAcee/UNPAN007872.pdf
Seidelman, R. (1985) Disenchanted Realists: Political Science and the American Crisis,
1884–1984, Albany, NY: State University of New York Press.
Waldo, D. (1948) The Administrative State: A Study of the Political Theory of American
Public Administration, New-York: The Ronald Press Company.
Weexsteen, A. (1999) Le conseil aux entreprises (Texte imprimé): le rôle de Jean Milhaud
(1989–1991) dans la CEGOS et l’ITAP, PhD thesis, Paris: École des Hautes
Études en Sciences Sociales.
Wilson, W. (1887) ‘Study of public administration’, Political Science Quarterly
2(2), 197–222.
46
THREE
The emergence of modern policy

analysis in France
Fabrice Hamelin
Introduction
This chapter focuses on policy process studies as they emerged and developed
in France from the 1950s to the early 1980s. During this pivotal period, the
conditions for their creation were favourable. Their establishment within academia
became clearly visible in the late 1980s when the first French manual of public
policy analysis was published by Mény and Thoenig (1989). The concluding
remarks of a review of the manual published in the Revue française de science
politique (Baudouin, 1990) asked the following question: ‘Ultimately, can one claim
that a new discipline has emerged?’ The response to this question was provided
following the review of the same manual in the Politiques et Management Public
(PMP) journal that had been created a few years earlier: ‘Policy analysis has gained
respectability. It has now become a Themis.’ Themis is a reference collection
by French universities and its acceptance by French universities marked a major
step forwards. Many other milestones were also achieved during this decade: the
creation of the first French journal focusing on public policy (PMP) in 1983; the
creation ex nihilo of a public policy laboratory by the French National Centre for
Scientific Research (the Public Policy Analysis Group) in 1984; the publication of
an entire volume of Political Science essays (Grawitz and Leca, 1985) focusing on
the analysis of public policies; the aforementioned manual by Mény and Thoenig,
and Pierre Muller’s (1990) Que-sais-je? which focused on the cognitive approach
to public policy. These markers of institutionalisation reflected the beginning of
the integration of policy process studies within French academic research, as an
autonomous ‘branch’ of political sciences compared to other policy approaches
developed in economic and legal disciplines.
To better understand how the establishment of policy process studies
within academic research has taken shape, it seems necessary to trace the
institutionalisation trajectory. This trajectory developed within the academic world
but also largely outside it, alongside it and in direct interaction with it. Indeed,
the institutionalisation of specific academic knowledge centred around ‘policy
process studies’ while ‘policy analysis’, that is, the knowledge produced and the
methodological tools developed to collect and analyse this knowledge in order
to respond to public issues, remained largely the product of state expertise. The
traditional clout of central government in France and the influence acquired by
47
the executive and its technocracy since the establishment of the Fifth Republic
in 1958 provide essential data for an understanding of how knowledge and the
methods that focus on understanding and controlling public policy in France
have developed. To better understand and present this unique development in
which researchers and innovative high-ranking government officials interact, it
appears necessary to outline the key phases.
The first phase began in the 1950s and arose from planning and national
accountability (Fourquet, 1980) as vectors of the rationalisation of public
policy. This was further developed in the 1960s thanks to the transfer of debates
and tools developed overseas, such as the rationalisation of budgetary choices
(Rationalisation des choix budgétaires [RCB]) inspired by the American PPBS. It was
not until the second phase, which commenced in the 1970s and was characterised
by the crisis of the welfare state and the transfer of policy studies knowledge
obtained by American universities, that expectations of the development and
institutionalisation of public policy analysis in academic research began. For the
actors behind this development, the first congress of the French Association of
Political Science held in the autumn of 1981 marked a major milestone. The
first roundtable on ‘public policy analysis’ was held and facilitated by those who
are today key figures in the development of the sub-discipline (J. Leca, J.L.
Quermonne, J.-C. Thoenig, among others) (Muller, 2008). This event marked
the establishment of public policy analysis within political sciences.
The transfer of decision-making tools

For two decades after the Second World War, social sciences were perceived
as insignificant within French universities. In France, reflections revolving
around public intervention were dominated by the action-oriented practices
and discourses of state experts, rather than by the analysis of these interventions.
Public policy analysis primarily sought the improvement of government policies
through the development of tools to rationalise them.
In the beginning was the Plan…
In the decade following the end of the Second World War, rapid growth
was observed in political-administrative thought within economic research
departments working on the costs and benefits of public intervention. The studies
undertaken sought to better understand economic developments and predict their
evolution. Moreover, they aimed to organise, or better organise, the institutional
and budgetary means at the service of collective decision-making. Consequently,
expert knowledge oriented toward decision-support tools developed within the
state apparatus and, specifically, within the three central institutions that played
a key role in the attempts to reboot the economy and rebuild French society.
Two institutions acquired the authority to act thanks to the massive economic
support proposed at the time by the Marshall Plan. Incentives proposed included
48
The emergence of modern policy analysis in France
the State Planning Commission (CGP) and a department of the Ministry of

Finance – the service of economic and financial studies (SEEF). Created in
1952 within the French treasury, SEEF was a national accountability department
responsible for establishing the accounts and budgets of the nation. It was involved
in developing the Plan and it undertakes economic forecasting. The action of
these two institutions thus shaped the French economy, at least in part, as the
United States chose to target the sections that received aid. However, according
to high-ranking government officials in these two institutions, planning was at
the origin of the reflection on the formulation of government policies and the
social consequences of government intervention (Spenlehauer, 1998). There was a
drastic change in the situation in 1953 when the funding granted by the Marshall
Plan finally came to an end. One of the consequences was the development and
sophistication of the national accounting system. In both the Ministry of Finance
and the entourage of the Plan, an institutional concern was observed in relation
to the implementation of economic calculations and cost–benefit analyses for
public policy.
This tropism was also observed within the Deposits and Consignments Fund
(Caisse des Dépôts et Consignations [CDC]). This fund enjoyed the use of the
savings of the general public and operated based on a public policy apprehension
model that differed from the CGP model. Less concerned with coordinating
state intervention in the various sectors, the CDC further embodied a highly
interventionist state model, albeit one in which policies were independent of
each other. It thus acted within the fields of housing, urban planning and even
in highway construction. The model proposed by the CDC thus appears to be
more in line with a growing economy compared to a global intervention expected
to revive production.
In 1965, the SEEF became a fully-fledged department: the forecasting
department. The decree of 9 July 1965, confirmed that forecasting would hence be
part of the traditional SEEF activities (participation in the preparation of national
accounts, responsible for economic budgets, participation in the development
of the Plan). The SEEF, however, was also given new responsibilities relative
to economic analysis and in relation to what was being done abroad. With
prospecting and analysis as its missions, the new department was responsible for
improving economic decision-support tools. In particular, it sought to develop
microeconomics and rationalise budgetary choices.
RCB was clearly modelled on the American PPBS implemented by the
Johnson administration. Following study visits to leading universities and tracking
missions in the United States, high-ranking officials of the budget and forecasting
departments set up a technological watch on PPBS at the Embassy of France in
Washington (Spenlehauer, 1998). Returning missionaries were convinced that
PPBS went beyond simply rationalising public spending. It could also be used as
a tool to reform public decision-making in all policy areas. The same was true
for another government technology in the United States, that is, the evaluation
of public policy (Spenlehauer, 1998). The issue of the harmonisation of the
49
economic and social modernisation of France affected the relationship between

the Ministry of Finance and line ministries alone. It was required of the entire
state apparatus and aimed to modernise public administration.
Having given rise to several study trips and missions in which senior French
officials visited the United States, the RCB thus became a major issue of the new
expertise that took shape within the central apparatus of the state in the 1960s
and 1970s. A sub-directorate of the RCB was thus created within the forecasting
department of the Ministry of Finance in 1968. This movement occurred in
parallel with the emergence of a new generation of young experts, economists
and engineers of major state institutions, well versed in the use of mathematics
and econometric tools (Benamouzig, 2005). The RCB sought to change the
budget development rules by transforming input-based budgeting, renewed
annually, into programme budgeting. However, its objective appeared much more
limited than the US model as it focused on ad hoc studies rather than on system
analyses. Expected to ‘rationalise’ policy decision-making by mathematically
calculating the economic and social efficiency of the contemplated or driven
policy, the RCB was extended to various ministries, notably social, such as the
Health ministry in which the analysis of the perinatal period acted as a reference
for many other administrations (Benamouzig, 2005). The same was observed in
the transport sector where the RCB was experimented with in the field of road
safety (Kletzlen, 2007, 139–46) as well as in the housing sector (Zittoun, 2001).
Then came the renewal of the framework for action following the institutional
innovations of the Fifth Republic
The state apparatus in France became the main venue for the development of
innovative tools for the management and analysis of public policy. It is necessary,
however, to go beyond ministries’ research departments to fully understand this
apparatus. A key factor in its development lay in the change of political regime
in 1958. The Fifth Republic, in particular the election of the President of the
Republic by direct universal suffrage in 1965, gave prominence to the executive.
Planning was raised to the rank of ‘major government obligation’, and in 1961
the CGP began to report directly to the Prime Minister. In response, the Finance
Minister, Valéry Giscard d’Estaing, transformed the SEEF into a forecasting
department. The longer ministerial mandates and the stability and expansion of
their staff also meant that administrations and the manner in which they were
managed also received more specific and sustained political attention. Although
the example of the RCB study on the perinatal period was considered one of
the most successful, it shows that the office of the Minister of Health did not
limit itself to the logics of econometric rationalisation alone as it also freely used
the results of the study to account for the different choices it made with regard
to more traditional policy variables (Benamouzig, 2005).
Thus developed, although decision-support tools explained decisions, they fell
within a much broader set of decision-making factors. This explains why the
50
operations undertaken by the RCB experienced a rapid decline in the 1970s.

Among the various explanations, the isolation of this expertise in relation to
the operational services of the administration must be noted. However, the
RCB methods and econometric tools survived the decline by participating in
administration at less strategic but more operational levels and engaging in more
ad hoc initiatives.
The political focus on expert analyses was also linked to the changes in French
society during the period. As Jean-Claude Thoenig (2008) has stated, one of the
first approaches resulted in the increase in the cognitive abilities of ministries.
Turning once again to services and consultants, the state and its ministries equipped
themselves with tools to better understand social change. The second approach
called upon academics even though intellectual and financial ties already existed
between planners and sociologists.
A key example is provided by the authorities responsible for urban planning
and development who, in the 1960s and 1970s, launched tenders focused on
the social sciences. There was a more than fivefold increase in the funding for
‘urban research’ between 1969 and 1976. The programme managers in the
relevant ministries thus established some form of complicity with a group of
researchers whose emergence they fuelled, and who were largely dependent on
them.From the mid-1960s, high-ranking planning officials considered that French
planning could no longer rely on economics alone; it also needed sociologists.
Sociologists were thus requested to help economists specialised in planning better
understand the expectations of French society as well as the resistance to change.
The objective of these modern senior officials trained in French Grandes Écoles
(École Nationale d’Administration, École Polytechnique) was to reconcile economic
development and the legitimate aspirations of the French in a changing society
(Spenlehauer, 2004, 124). They sought to better understand and interpret the
socio-economic environment of France, as well as its developments, in order to
envisage the orientations and major collective choices of the nation.
For example, the preparatory studies of the RCB relating to the perinatal period
and to road safety were no longer exclusively inspired by economics; other social
sciences were also integrated. Similarly, this was perceived in the orders placed for
planning activities in the early 1970s. Although these concerns raise questions on
the low impact of the planning apparatus on public policy, they were the result
of the government’s will to understand and integrate new social and qualitative
concerns revealed during the 1968 crisis (Jobert, 1995, 14). This awareness was
boosted in the wake of the shock provoked by the events of May 1968 which
made the need to better understand society and to increasingly rationalise the
decision-making process all the more legitimate. To understand how French
public policy analysis developed in the 1960s, one must thus take into account
the state’s – and its high-ranking officials who spoke on its behalf – attempts to
adapt to French society. Following the social and political shock provoked by the
events of 1968 and even before the economic crises of the 1970s, the desire to
understand a society in upheaval and re-examine state intervention were further
51
objectives that drove research procured by government. Within a centralised yet

deeply segmented French government, dedicated offices and research centres
were created, giving rise to highly significant sectoral policy analyses (Le Galès,
2015). We can cite, for instance, the development of a Marxist-inspired critical
analysis of urban policies that notably shifted the focus toward policies’ target
populations. This desire also reflects research sponsors’ acceptance of a critical
analysis of state action and their openness to new frames of reference, that is,
those from the academic world and from abroad. This was undoubtedly one of
the key contributions of the development of public policy analysis in academic
research in the 1970s.
Academic institutionalisation
In the 1960s and 1970s, public policy studies emerged and developed in the
academic setting. Although these studies were developed in various areas, they
remained rare in France (Muller 2008). Consequently, the development of public
policy analysis in academic research was often limited to the studies undertaken
in two higher education institutions, one situated in Paris and the other in the
provinces: the Centre for the Sociology of Organisations (Centre de Sociologie des
Organisations [CSO]) and the Centre for the Analysis of Regional Administration
(CERAT) in Grenoble. This tended to overshadow the studies conducted
elsewhere (Le Galès, 2015) and influenced by other analytical trends – such
as Foucault’s work for example – which could further develop the knowledge
on specific sectoral policies (social, urban and legal policies in particular). As a
result, the differences between these two main areas of knowledge production
were accentuated and the diversity of the studies undertaken was overlooked.
Moreover, the studies carried out abroad, for instance in the United States and
in other parts of Europe, were underestimated.
In the beginning was public procurement…
Within universities, the first policy studies were conducted by economists and
lawyers. Political science was perceived simply as a sub-discipline for publicists and
constitutionalists. Until the early 1970s, public policy analysis was thus associated
with ‘administrative sciences’ in law schools and with ‘political economy’ in the
faculties of economics (see Gaudin, 2004). Moreover, the Institutes of Political
Studies (Instituts d’Etudes Politiques [IEP]), which were an eclectic branch of
universities, did not favour political sciences, despite the fact that this academic
discipline was first structured around the IEP of Paris, the National Foundation of
Political Science and the French Political Science Association. Research activities
were focused on Paris, but they were primarily oriented toward the study of
the political activities and behaviour (voting, political organisations, and so on).
The encounter between the high-ranking officials responsible for planning,
political science and, more broadly, French social sciences was not all smooth
52
sailing during this period (Spenlehauer, 2004). There were, however, spaces and
opportunities for exchange. The CGP and the Ministry of Finance initiated orders
to assess how the Marshall Plan aid funds were being managed. Their objective
was to analyse reconstruction and development policies. They also sought to
develop descriptive models and policy analysis grids and, naturally, to examine
the administration’s weak points (Gaudin, 2004). Informed senior officials also
worked on building lasting ties between government and researchers, as we
previously highlighted in the case of urban policies.
In the late 1950s, discussions were held with a small group of academics, notably
sociologists, with Michel Crozier as one of the participants (Gaudin, 2004). When
Crozier returned from a research stay in Stanford, he created the CSO in 1961.
The rationality of public decision-making was discussed and challenged because
of its dysfunctions. The focus on government and on political elites was also
crucial. The falsely centralised and highly fragmented state was simply perceived
as a system of actors and actions. It was thus necessary to open the black box of
the political-administrative apparatus in order to develop a strategic analysis of
policy-making. This sociological approach expressed the loss of confidence in the
ability of the state to solve economic and social problems. The main intellectual
influence was methodological individualism which used a comprehensive
approach to focus on actors and their impeded rationality (Ledoux and Pollard,
2015). This was illustrated by the use of monographic and qualitative approaches.
The influence of CSO cannot be reduced to the reputation and the studies
of Crozier alone. The studies undertaken by other researchers, from a different
generation, including Jean-Claude Thoenig, Pierre Grémion and Jean-Pierre
Worms, are also worth mentioning. This underscores the diversity of the
professional trajectories pursued by those who were fortunate enough to pass
through this laboratory. For instance, Thoenig left CSO in 1974 to join the
Polytechnic of Lausanne, followed by the IEP of Bordeaux and, especially, Institut
Européen d’Administration des Affaires (INSEAD) in Fontainebleau. Drawing
on the contributions of organisational sociology to understand public policy no
longer sufficed. The turn toward other educational institutions indicated the shift
from the internal analysis specific to CSO toward the analysis of ‘interactions’
and inter-organisational relationships around which Thoenig’s studies on the
sociology of public policy focused (Le Galès, 2015). They also explained Crozier’s
lack of interest in public policy analysis. However, Thoenig founded the Public
Policy Analysis Group (Groupe d’Analyse des Politiques Publiques [GAPP]) in 1984,
having left CSO over a decade earlier. Symbolically, the first French research
centre dedicated to public policy analysis was created by the CNRS in Cachan,
within the École Normale Supérieure. This was a bona fide department rather than
a joint research unit. This marked the intellectual independence of public policy
analysis and political science, as evidenced by the creation of Section 40 as one
of the CNRS sections.
The establishment of public policy analysis within the political sciences was even
greater in CERAT. The research undertaken in CERAT under the direction of
53
Lucien Nizard drew much more heavily on the studies by Marx and Durkheim
(the social division of labour and political regulation). These studies expressly set
themselves apart from the systemic approach used by the researchers trained at
CSO, an approach considered inadequate from a political perspective. For instance,
Jobert (1995) denounced ‘the recurrent denial of policies professed by researchers
specialised in the sociology of organisations’. The researchers trained in CERAT
in the 1970s were not ‘on the same wavelength’ as those trained at CSO. There
were, therefore, genuine points of divergence between the two laboratories.
The relationship between the state and society primarily drew on planning
studies, as evidenced by the ‘Planning and Society’ conference held in Uriage in
1973 or, more broadly, by the studies undertaken by Nizard (1974) and by Jobert
in his thesis on urban planning (Jobert, 1977). They were also based on sectoral
policy analyses (agricultural sector, social sector, and so on) and the analysis of
programmes (Airbus). The key objective was to analyse how the emergence
of new policies was connected with wider global trends. Analysing planning
validated the centrality of state actors (elected members, high-ranking officials
and experts). However, this analysis also underscored the role played by interest
groups in the implementation of public policies and revealed the inequalities that
shaped the relationships between social groups. The study of planning made it
possible to highlight how different sectors influenced public policies and to reflect
on how these sectors and, more generally, French society, viewed themselves.
Planning was perceived as a tool for disseminating a shared vision of the future of
a society (state transformation and modernisation) as well as a means of reducing
distortions between projects designed and driven by independent actors. The
planning triangle that connected the state planning commission, the forecasting
department of the Ministry of Finance and Institut National de la Statistique et
des Etudes Economiques (INSEE) produced a worldview that revolved around
modernisation (Payre and Pollet, 2013, 26). The researchers who conducted their
PhD studies at CERAT in the 1970s thus distinguished themselves from structural
Marxism by empirically challenging an overtly monolithic view of both social
domination and the state. The studies undertaken paid special attention to the
role of ideologies, professional groups, senior civil servants and neocorporatism
(Jobert and Muller, 1987).
The theoretical bricolage, facilitated by the analysis of the empirical phenomena
observed, involved the introduction and the reformulation or distortion of
traditional concepts, such as the concept of ideology. These ‘organic intellectuals’
expressed the worldview held by social groups. According to Thoenig (2008),
while this reflected a holistic approach that might also be described as ‘hyper-
deterministic’, it was also a top-down approach. These findings must be treated
with caution, however, because they cannot be indiscriminately applied to all
the studies conducted at CERAT, where several generations of researchers are
still at work. Nevertheless, these critical issues were published in L’état en action
in 1987. In 1995, however, Jobert rejected the accusations levelled against him
with regard to transmitting, notably through the référentiel, a global and elitist
54
vision of public policy. He argued that this approach, which was interactionist
above all, had been misunderstood as the first synthesis report had undoubtedly
presented public policy in far too schematic a manner.
While this type of debate referred to different policy-making approaches and to
the use of distinct theoretical tools, the studies carried out in the two laboratories
nonetheless shared some similarities. They distanced themselves, on the one hand,
from the legal concept of policy analysis and, on the other, from the dominant
vision of the state as a bureaucratic machine in the service of society, or as an
instrument that dominates society. In the studies undertaken at both CERAT
and CSO, the state was initially presented as a relatively unstable assemblage
of services and public policies with problematic rationality. Both laboratories
underscored the contradictions between states’ various programmes and public
policies, devoid of consistency, divided and segmented. Although these points
of convergence resulted from different theoretical corpus, they were reinforced
by the working relationships forged via the Délégation interministérielle à
l’aménagement du territoire et à l’attractivité régionale (DATAR) and its ‘political
foresight’ programme (Jobert, 1995, 15). Moreover, the research contexts of the
two laboratories were quite similar as they were structured by the CNRS and the
IEPs. Initially, those behind policy analysis sought to transform this analysis into
a branch of political science receptive to other sub-disciplines within the political
sciences. Public policy analysis, which drew on comprehensive and qualitative
studies, was developed at the crossroads of sociology and administrative science
studies.
Field studies and sectoral policies were prioritised, as evidenced by the content
of Volume 4 of the political science treatise (Grawitz, Leca, 1985) introduced by
Thoenig (1985), and Padioleau’s Concrete State Structures (L’Etat au concret) (1982).
They were also a priority in the theses undertaken in Grenoble and published
in the early 1980s – Le Technocrate et le paysan (The technocrat and the peasant)
was published in 1984 – which served as a foundation on which to shape the
key concepts of the cognitive analysis of public policies. It must be noted that
the focus on field studies arose from the fact that public orders were behind the
cases analysed, thereby giving public policy analysis a strong applied dimension.
Although methodological concerns were not at the centre of the studies
undertaken, the practices analysed could be ‘diverted’ for purposes of scientific
theorising. The desire to develop scientific theories gave the emerging sub-
discipline autonomy and a scientific nature, as evidenced by Muller’s reflections
around action ‘reference points’ and Thoenig’s reflections on ‘crossed regulation’.
The objective of creating a ‘toolbox’ rather than a ‘how to’ manual to more
effectively deconstruct public policy was to make public policy analysis
independent. To this end, state and power theories were reintegrated into this
analysis, to the benefit of policy specialists (Thoenig, 1996). French public policy
analysis within the academic setting thus took shape in response to administration
sciences acquired from public law. It was also influenced by concepts developed
overseas, particularly those developed in North American universities.
55
…then came the transfer of policy process studies
This period was also marked by the internationalisation of policy process studies, a
process in which French researchers were active participants. Under the direction
of Crozier, the CSO largely contributed to the transfer of American reflections,
through Crozier’s networks and studies, as well as his close ties with American
research institutions and foundations which facilitated the transatlantic exchange.
Among the transferred concepts were bureaucratic dysfunctionalism, bounded
rationality, trial and error and bricolage, drawn from the studies of R. Merton,
H. Simon and C. Lindblom. The objective was to question the content of public
policy programmes, the underlying causes of the problems and the origins of
proposed solutions. This, therefore, was an academic approach rather than a
simple operations-oriented analysis. Import–export studies from the United
States explain why public policy analysis emerged as a subversive discipline in
which state action was criticised by the members of major public institutions.
To better understand this influence, the studies by researchers who undertook
policy analysis based on field studies and implementation must also be mentioned;
these include the pioneering studies of Dahl, as well as the studies undertaken
by Pressman and Wildavsky (1973) and Gusfield (1980). The academic venues
in which these studies were conducted, such as the universities of Berkeley,
Stanford, Cornell and Harvard, are also worth mentioning. Reference studies
were first characterised by the development of sectoral analyses – the pioneering
studies of Dahl (1961) and Pressman and Wildavsky (1973) – which explains why
French studies focused on both decision-making processes and public policy
implementation. For French researchers studying in the United States, these
studies, unheard of in France, thus came as a surprise. Padioleau and Thoenig
perfectly embodied the image of the 1970s ‘importing-traveller’. Others soon
joined them, repeatedly making the transatlantic journey: including Y. Mény, J.
Leca, J.L. Quermonne. Stays in American universities presented an opportunity to
read what was being done in the country at the time; it was a time for discovery
as well as for encounters and interpersonal exchanges. Beyond the transfer of
concepts, researchers read and discussed with other researchers working in North
American universities. Despite this North Atlantic flow, however, some theoretical
approaches (rational choice, the instrumental vision of public policies around
problem resolution) were not transferred to France.
Naturally, not all scholars of this generation took part in this movement. Indeed,
‘importers’ able to read English were few in number. Moreover, the excessive
emphasis of authors such as C. Jones suggests that the Anglo-Saxon field was
poorly controlled. Gaudin (2004, 16) believes, however, that the development
of French public policy analysis was not really under-developed compared to the
United States, or possibly slightly under-developed, like all the other fields of
political science. Others have supported this interpretation. Jobert, for instance,
argues that real progress in French public policy analysis only occurred when
56
French researchers began to participate in international debates and activities

(interview held on 29 May 2015).
The United States thus made important contributions to French public policy
analysis. Other parts of Europe also influenced this policy, notably Germany
and Britain. The import–export process was thus also European. With regard
to the studies undertaken jointly with British (Hayward, 1974) and German (G.
Lehmbruch) researchers, Jobert has given priority to the studies undertaken in
Grenoble, notably those carried out in the context of comparative policy analysis.
This view is shared by Thoenig, who mentions the influence of the studies of
D. Ashford in Britain and R. Mayntz in Germany (interview held on 3 June
2015). Public policy analysis was also emerging elsewhere in Europe. Indeed, the
stays in the United States did not mean that French and American researchers
were exclusive partners. Other European researchers also participated in these
transatlantic exchanges, including the aforementioned Renate Mayntz. French
public policy analysis thus developed in a favourable context in which young
researchers participated in transatlantic trips at a time when American policy
analysts were turning away from systemic analyses and renewing the public
policy analysis toolbox. These trips were made possible through scholarships – a
post-doctoral fellowship from the Ford Foundation for Thoenig – which thus
structured the studies carried out overseas and influenced the establishment of
European public policy analysis.
The founders of this academic discipline were well versed in international
literature, and this explains why French research was not overtly singular. In 1990,
referring to the manual by Mény and Thoenig, Gibert thus observed that ‘the
authors are knowledgeable in international literature and this allows them to avoid
all attempts to reinvent the world or propose a policy study à la française which
is unnecessary’. However, this French policy process study, actively involved in
a much wider transnational movement, was long overlooked in English public
policy analysis journals (Thoenig, 1996). This reflected the infancy of the sub-
discipline, the difficulties encountered by some researchers in mastering English,
and the uncertain international recognition and singular orientation associated
with the early studies written in French. This singular tradition is reflected, for
instance, in the emphasis placed on the relationship between public policies
and regions (local and European regions under construction), and more so in
the development of cognitive approaches in public policy through the focus on
controversial policies and on translation issues, as well as on the emphasis placed
on contractual policies (Muller, 2008) and sectoral policies (social and urban
policies in particular).
The transatlantic travels opened up new research activities and also challenged,
or at least made it possible for researchers to distance themselves from, the studies
carried out in France, notably in law and administrative sciences. How, then, did
French researchers nourished by American policy analysis studies and French
intellectual traditions respond? French sociologists and political scientists have
long studied public policy by analysing the state, administrative science and several
57
public policies. What remains of these initial questions? The question is a tough
one to answer, because no inventory of the contributions of studies inspired by
the legal land administrative science fields exists (Payre and Pollet, 2013, 24).
However, the renewal of studies on the state in action paradoxically led to the
exclusive development of public policy analysis as a sub-discipline of political
science, in contrast to the United States (Massardier, 2003, 15).
The institutionalisation of French policy studies might also be explained by
the traditional competition between interested academic disciplines. Economists,
sociologists and political scientists have always competed with one another to
analyse the bureaucratic phenomenon and certain public policies, such as urban
policies. The focus on the constraints faced by politicians and, in particular, the
role played by politics in the policy-making process (Muller, 1996, 98) is one
of the issues that fall within this game of interdisciplinary distinction. However,
in the late 1970s the establishment of policy studies within political science was
not always straightforward. For instance, one criticism levelled against Mény and
Thoenig’s manual was that it perceived economists as adversaries who needed
to be excluded from the explanatory field. The most significant issue, however,
was the gap in the political science in the French academic setting itself. Some
of its members, who were the most legitimate at the time, discounted emerging
studies on state action. ‘Traditional’ policy analysis was more interested in how
policies influenced society and, thus, the state, and it focused on inputs (electoral
attitudes and behaviour, political participation) rather than on outputs. ‘Election
analysts’ thus dominated the academic scene. The same was observed in the
historical sociology of the state, which disregarded public policies, such as the
Sociologie de l’État by Pierre Birbaum and Bertrand Badie published in 1979. This
phenomenon was further reinforced by the fact that practitioners, notably high-
ranking officials, including those who taught at Sciences Po, remained insensitive
to the studies undertaken by social scientists, even if there are a few exceptions
(De Kervasdoué et al, 1976). The main interlocutors of public policy analysts
were members of the community of political scientists, and the legitimacy of
approaches was dependent on their scientific nature and on peer recognition. As a
result, policymakers distanced themselves from practitioners and decision-makers,
who were more interested in social debate and decision-support (Muller, 2008).
Alongside the impact of academic competition thus lay the ignorance of social
science studies among high-ranking officials who had remained more attentive
to law and economics than to political science.
Conclusion
It is possible to draw some important conclusions from the initial phase of the
institutionalisation of public policy analysis in France. Lecturers and researchers
were not alone in looking elsewhere – in the United states and in Europe – for
concepts that nourished policy process studies. The impact of the missions
undertaken by modern high-ranking officials, which led to the development
58
of the RCB and the evaluation of public policies within the state apparatus
and, therefore, at the margins of the university, must not be overlooked. The
researchers and academics who also developed public policy analysis in France and
contributed to its institutionalisation within the academic setting first attempted
to make it a branch of the political sciences. Put differently, they attempted to
make it a scientific (sub-)discipline in its own right, at the risk of isolating it
from the shift toward the rationalisation of state action. This institutionalisation
was thus shaped by multiple forces that influenced its development: political
decentralisation, the questioning of the welfare state, and so on. The public finance
crisis, decentralisation and European integration helped ‘loosen the state’s grip’ on
society. These changes also gave way to newly emerging needs: decentralisation
led to an influx of high-ranking officials into local authorities. The French civil
service centre (CNFPT) responsible for training local government employees
thus integrated the analytical contribution of social sciences to the management
of public policy programmes into its training curriculum. This, however, was not
the case in the state’s administrative apparatus (Thoenig, 2008).
These different movements make it difficult to assign one particular meaning to
the generated momentum. The image of organised anarchies is thus not a totally
exaggerated explanation. However, French history allows us to underscore the
significance of empirical sectoral analyses and the diversions used by researchers
to achieve the generality needed to prove that the studies produced are of a
scientific nature. It is therefore worth investigating whether the theoretical limits
and weaknesses of these diversions, much like the divide that this momentum
installed between the world of public managers and scientific knowledge, led to
the collapse of public policy analysis à la française conducted within university
settings in the 1990s (Hassenteufel and Smith, 2002).
References
Badie, B. and Birnbaum P. (1979) Sociologie de l’État, Paris: Grasset.
Baudouin, J. (1990) ‘Mény Y., Thoenig J.C.: Politiques publiques’, Revue française
de science politique 40(3), 394–7.
Benamouzig, D. (2005) La santé au miroir de l’économie: Une histoire de l’économie
de la santé, Paris: Presses universitaires de France.
Boussaguet, L., Jacquot, S., Ravinet, P. (eds) (2015) Une ‘French touch’ dans l’analyse
des politiques publiques?, Paris: Presses de Sciences Po.
Dahl, R. (1961) Who governs? Democracy and power in an American city, New Haven:
Yale University Press.
De Kervasdoué, J., Fabius, L., Mazodier, M., Doublet, F. (1976) La loi et
le changement social: Un diagnostic. La loi du 16 juillet 1971 sur les fusions et
regroupements de communes, in Revue française de sociologie 17(3), 423–50.
Dubois, V. (2002) ‘Socio-histoire et usages sociaux de l’histoire dans l’analyse de
l’action publique: Réflexions à partir de la politique culturelle en France’, dans
Y. Déloye, B. Voutat (eds) Faire de la science politique, pp. 155-69, Paris: Belin.
59
Dubois, V. (2009) ‘L’action publique’, in A. Cohen, B. Lacroix, Ph. Riutort (eds)

Nouveau manuel de science politique, pp. 311–25, Paris: La Découverte.
Dupuy, F., Thoenig, J.-C. (1983) Sociologie de l’administration française, Paris:
Armand Colin.
Dupuy, F., Thoenig, J.-C. (1985) L’administration en miettes, Paris: Fayard.
Fontaine, J. (1996) ‘Public policy analysis in France: Transformation and theory’,
Journal of European Public Policy 3(3), 481–98.
Fourquet, F. (1980) Les comptes de la puissance: Histoire de la comptabilité nationale
et du plan, Paris: Encres, Éditions Recherches.
Gaudin, J.-P. (2004) L’action publique: Sociologie et politique, Paris: Presses de
Sciences Po and Dalloz.
Gibert, P. (1990) ‘Yves Meny et Jean Claude Thoenig: Politiques Publiques,
Politiques et management public 8(1), 155–6.
Grawitz, M., Leca, J. (eds) (1985) Traité de science politique, vol 4, ‘Les politiques
publiques’, with a presentation by J.-C. Thoenig, ‘L’analyse des politiques
publiques’, Paris: Presses universitaires de France.
Gusfield, J. (1980) The culture of public problems, Chicago: The University of
Chicago Press.
Hassenteufel, P., Smith, A. (2002) ‘Essoufflement ou second souffle? L’analyse
des politiques publiques “à la française”’, Revue française de science politique, 52(1),
53–73.
Hayward, J. (1974) ‘National aptitudes for planning in Britain, France and Italy’,
Government and Opposition 9(4), 397–410.
Hayward, J., Narkiewicz, O.A. (eds) (1978) Planning in Europe, New York: St
Martin’s Press.
Jobert, B. (1977) La planification urbaine: l’exemple de Stuttgart, Grenoble: Presses
Universitaires de Grenoble.
Jobert, B. (1992) ‘Représentations sociales, controverses et débats dans la conduite
des politiques publiques’, Revue française de science politique 42(2), 219–34.
Jobert, B. (1995) ‘Rhétorique politique, controverses scientifiques et construction
de normes institutionnelles: esquisse d’un parcours de recherche’, in A. Faure,
G. Pollet, P. Warin (eds) La construction de sens dans les politiques publiques. Débats
autour de la notion de référentiel, pp. 13–24, Logiques politiques, Paris: L’Harmattan.
Jobert, B., Muller, P. (1987) L’État en action: Politiques publiques et corporatismes,
Kletzlen, A. (2007) De l’alcool à l’alcool au volant, Paris: Collection Logiques
sociales, L’Harmattan.
Kübler, D., Maillard, J. de (2009) Analyser les politiques publiques, Fontaine: Presses
universitaires de Grenoble.
Le Galès, P. (2015) ‘Les approches françaises des politiques publiques au temps de
Pierre Muller. Complément d’enquête’, in L. Boussaguet, S. Jacquot, P. Ravinet
(eds) Une ‘French touch’ dans l’analyse des politiques publiques?, pp. 185–201, Paris:
Presses de Sciences Po.
60
en France’, in O. Giraud, P. Warin (eds) Politiques publiques et democratie, Paris:
La Découverte.
Ledoux, C., Pollard, J. (2015) ‘L’héritage intellectuel de l’analyse des politiques
publiques’, in L. Boussaguet, S. Jacquot, P. Ravinet (eds) Une ‘French touch’
dans l’analyse des politiques publiques?, pp. 203–20, Paris: Presses de Sciences Po.
Massardier, G. (2003) Politiques et action publiques, Paris: Armand Colin.
de France.
Muller, P. (1984) Le technocrate et le paysan, Paris: Les éditions ouvrières.
Muller, P. (1985) ‘Un schéma d’analyse des politiques sectorielles’, Revue française
Muller, P. (1990) Que sais-je?, Paris: Collection, Presses universitaires de France.
Muller P. (1996) ‘Cinq défis pour l’analyse des politiques publiques’, in P. Muller,
J. Leca, G. Majone, J.C. Thoenig, P. Duran (eds) ‘Enjeux, controverses et
tendances de l’analyse des politiques publiques’, Revue française de science politique,
46(1), pp. 96-102.
Muller, P. (2008) ‘Analyse des politiques publiques et science politique en France:
“Je t’aime, moi non plus”’, Politiques et management public 26(3), 51–6.
Muller, P., Leca, J., Majone, G., Thoenig, J.-C., Duran, P. (1996) ‘Enjeux,
controverses et tendances de l’analyse des politiques publiques’, in Revue française
Nizard, L. (ed) (1974) Planification et Société, Fontaine: Presses universitaires de
Grenoble.
Padioleau, J.G. (1982) L’Etat au concret, Paris: Presses universitaires de France.
Payre, R., Pollet, G. (2013) Socio-histoire de l’action publique, Paris: La Découverte.
Pressman, J.L., Wildavsky A. (1973) Implementation, Berkeley: University of
California Press.
Sfez, L. (1973) Critique de la décision, Paris: Presse de la FNSP.
Spenlehauer, V. (1998) L’évaluation des politiques publiques, avatar de la planification,
Grenoble: Humanities and Social Sciences, Université Pierre Mendès-France
– Grenoble II.
Spenlehauer, V. (1999) ‘Intelligence gouvernementale et sciences sociales’, Politix
12(48), 95–128.
Spenlehauer, V. (2004) ‘Pour une déconstruction des légendes sur les rapports
Etat-Sciences sociales’, in B. Zimmermann (ed) Les sciences sociales à l’épreuve de
l’action: Le savant, le politique et l’Europe, pp. 119–44, Paris: Éditions de la Maison
des Sciences de l’Homme.
Thoenig, J.C. (1985) ‘“Présentation” and “L’analyse des politiques publiques”’,
in M. Grawitz, J. Leca (eds) Traité de science politique. 4, Les politiques publiques,
p.I-X and 1-60, Paris: Presses Universitaires de France.
61
Thoenig, J.C. (1996) ‘La quête du deuxième souffle’ in P. Muller, J. Leca, G.

Majone, J.C. Thoenig, P. Duran ‘Enjeux, controverses et tendances de l’analyse
des politiques publiques’, Revue française de science politique, 46(1), pp. 102-107.
Thoenig, J.-C. (2008) ‘Politiques publiques et cycles de vie: Le bébé et l’eau du
bain’, Politiques et management public 26(3), 57–76.
Zittoun, P. (2000) ‘Quand la permanence fait le changement: Coalitions et
transformations de la politique du logement’, Politiques et management public
18(2),123–47.
Zittoun, P. (2001) La politique du logement, 1981–1995: Transformations d’une
politique publique controversée, Paris: L’Harmattan.
Zittoun, P. (2013) La fabrique politique des politiques publiques: Une approche
pragmatique de l’action publique, Paris: Presses de Sciences Po.
Major textbooks in policy studies published during the 1980s in

France
Jobert, B., Muller, P. (1987) L’État en action, politiques publiques et corporatismes,
de France, Themis.
Muller, P. (1990) Les politiques publiques, Paris: Collection ‘Que sais-je?’, Presses
Thoenig, J.-C. (1985) ‘L’analyse des politiques publiques’, in M. Grawitz, J. Leca
(eds) Traité de science politique, vol 4, pp. 1–60, ‘Les politiques publiques’, Paris:
Presses universitaires de France.
62
FOUR
Recent developments within

French policy studies
Jacques de Maillard and Andy Smith
Introduction
The previous chapter has just explained how policy studies emerged and developed
in France in the 1970s and 1980s. This one picks up the baton to show how, from
these humble but crucial beginnings, policy studies have since grown in France
and become such an institutionalised and omnipresent part not only of sociology
and political science, but also of expertise on public affairs. In so doing, two of
its key features need highlighting from the outset.
First, as a field policy studies in France has expanded partly because its
proponents have consistently allied studies of ‘new’ political phenomena (for
example, local and EU public policies) with the concerted development and
discussion of concepts and analytical frameworks (for example, governance,
networks, political leadership). It is to be noted that gender policy studies have
contributed to renewing these perspectives by opening new fields of studies (both
the gender of policies and the policies of gender) while at the same time discussing
classical notions (for instance the ‘referential’, see below) and perspectives (for
example, the study of street-level bureaucrats) (see the recent syntheses by Engeli
and Perrier, 2015; Mazur and Revillard, 2016). Although of course present in
other national fields, this linkage between empirics and theory is particularly strong
in France. For these reasons, we have chosen to present what follows around three
fundamental debates which, through developing in layers over the last 30 years,
structure and divide the French policy studies of today. The first concerns the
role of ‘ideas’ in public policies, the second the relationship between institutions
and actors, while the third is centred upon the role of the state and politics itself
within policy-making, implementation and evaluation.
Second, this theoretical investment has gone hand in hand with an ever-
increasing engagement with research published in English. However, this
involvement in international social science has nevertheless varied in intensity
and intentionality over time. If, in the 1990s, this field of study was essentially
autonomous from extra-national developments, by contrast the following decade
was marked by importation and translation of approaches initiated elsewhere
(Muller and Surel, 2000). Since the end of the 2000s, however, greater cross-
fertilisation between endogenous and exogenous perspectives has emerged,
63
alongside a greater willingness to participate assertively and cumulatively in

international scientific fora.
Throughout what follows, we will not only show that the approaches which
have dominated this field of study in France continue to differ from the worlds
of policy studies in other countries. We will also demonstrate that a particularly
strong connection between political science and sociology is the prime cause of
this difference. As will be highlighted, the three recurrent debates dealt with here
are all linked to the deep-seated reasons why academics in France generally refuse
to engage in the ‘problem-solving oriented’ ‘policy analysis’ that is so common in
most other countries. Indeed, overall, this chapter is as much about the building
and evolution of two professions and professional identities – those of political
scientists and sociologists – as it is about change in the theories, methods and
practices of their respective members.
The impact of ideas upon public policies

Generally speaking, French policy studies has been marked by a reluctance to
adopt rational choice approaches which, in turn, has had the deep consequence
of rendering the latter marginal in this country. Instead, analytical frameworks
giving primacy to the role of ideas, norms and values have occupied a central
role within the parts of academia interested in policy-making. More specifically,
a group of researchers based in Grenoble during the 1980s prioritised the role
played by ideas in public policies on policy studies’ agenda, a place from which it
has incited sustained debate first over the 1990s then, more sporadically, ever since.
L’Etat en action and the concept of ‘referential’
Through striving to collectively develop an all-encompassing reading of state

action, these Grenoblois argued that representatives of the state neither simply
responded to the demands of interest groups, nor worked consistently in favour of
the ruling class. Instead, as Jobert and Muller in particular set out when combining
insights from post-Marxism and organisational sociology (1987), a reasoning in
three stages was deemed necessary.
• Research needs to grasp the dominant systems of meaning which structure

public policies through becoming their respective ‘referential’, a key term
conceptualised as the codes and models of reference providing orientations,
stability and legitimacy to policy-making actors (1987, 47).A referential
never dominates a policy sector by chance, but rather because it is part of the
strategic activity conducted by certain actors to become hegemonic. Labelled
‘mediators’ by Jobert and Muller, these actors not only participate actively in
shaping the cognitive aspects of a referential, they also build an image and a
role for themselves. In short, ideas are not free-floating, rather they are firmly
64
Recent developments within French policy studies
attached to the positions and legitimacy developed by the actors who promote
them (Radaelli, 2015, 222).
• The connected concept of ‘global–sectorial relationship’ guides research to
studying how the referential that dominates a sector is affected by that which
over-arches policy-making in a polity at that point in time. The claim made
here is that these all-encompassing interpretations of the social and political
world have structuring impacts because they tend strongly to determine the
scope of possibilities which sectorial mediators consider are open to them.
Research conducted in this vein has had massive success in France, and this
among young researchers in particular. Moreover, through giving rise to several
collective reflexions (Faure et al, 1995; Boussaguet et al, 2015), it has participated
strongly in the very institutionalisation of policy studies in this country that first
occurred in the 1990s. In so doing, this approach has been successfully mobilised
in analyses of policy changes, especially various ‘turns’ experienced by the French
policy model (for example, a referential of modernisation led by the state in the
1960s, then the turn to neoliberalism in the 1980s).It is of course important
not to disconnect this development from other similar ones that have occurred
elsewhere in the world among social scientists specialised in public action. In
the US and Britain concepts such as ‘paradigms’, ‘narratives’ or ‘traditions’ have
also been used as a means of countering positivist approaches to public policy,
and that of many rational choice theorists in particular (Muller and Surel, 2000).
Indeed, many prominent foreign colleagues have praised the concept of referential
for underlining how it links policy analysis to that of legitimacy, as well as the
linkages between the macro and the meso it guides research to reveal (how a
change in a specific sector is bound to a more macro change) (Hall, 2015). This
said, within France itself ‘the referential approach’ initially inspired an essentially
national debate that not only divided its proponents from both more interactionist
organisational sociologists and more structuralist followers of Pierre Bourdieu,
but also isolated the field as a whole from international debates. In doing so,
this intranational debate also provoked a number of extensions and revisions
(Boussaguet and Surel, 2015).
Extensions: interactionist takes on the production of ideas
A creator of the referential approach, Bruno Jobert, was one of the first to
extend it by focusing more specifically upon the origins of ideas that become
dominant and the range of actors who participate in this process. In particular,
his conceptualisation of ‘forums’ as relatively autonomous spaces of thought and
debate, enabled Jobert to analyse more precisely the shift within public action from
a Keynesian global referential to one based on neoliberalism (1994). Rather than
see this displacement as the result of work carried out by mediators, he instead
highlighted a plurality of learning processes which occurred within scientific,
expert, journalistic and political fora. Analyses of agricultural policy change
65
(Fouilleux, 2003), or legislation to counter paedophilia (Boussaguet, 2008) have

since tested and refined this claim.
A second interactionist-inspired development of the referential approach has
sought to reconcile it with analysis in terms of strategic action. Rather than
opposing ideas and interests, this strand of research has picked up on Muller’s
constructivist assertion that all interests flow from processes of decoding and
recoding, and this in order to underline that the ideational dimension of public
action is omnipresent within the powering that determines its outputs and
outcomes. Consequently, researchers have stressed the strategic uses of ideas
as a weapon within institutional games: for instance, the EU Commission has
used neoliberal policy recipes in its negotiations with Member-states as a way of
defending its own organisational interests (Crespy and Ravinet, 2014).
Finally, a third interactionist extension of the referential approach has sought
to meld research on ideas and on ‘communicational rationality’. From this angle,
the ideas which affect public policy are seen as influencing not only how actors
interpret the world, but also as structuring the discourses they use to change it
by interacting with others. Here, discourse is not opposed to reality, it is seen as
constitutive of the latter. Consequently, importance is given to both the practical
knowledge (savoir faire) of actors, as well as to its ‘political’ dimension (that is,
knowledge that is not solely scientific but refers to ‘the public good’, see Durnova
and Zittoun, 2013; see also in English for a discussion of Risse’s deliberative logic
of governance, Quantin and Smith, 2013).
Revisions: ideas as the causes of policy change or reproduction
What has provoked the most controversy within French policy studies, however,
concerns the causal status accorded to ideas within policy-making. Here four
differing positions have been put forward.
The first is to address the question of causality with the empirical finding that,
at least within contemporary pluralist democracies, the ideas that tend to ‘win’
in policy-making are those that are sufficiently flexible or ‘soft’ to enable actors
to reach compromises or a consensus. More precisely, the success of soft ideas
stems from them being sufficiently ambiguous to enable different, and sometimes
contradictory, interpretations to be accepted by all concerned (Palier, 2003).
For example, this claim has subsequently been made to explain the success of
neoliberalism in so much of the world since the 1980s (Crespy and Ravinet,
2014), as well as the heterogeneity of its translations into practice at the national
scale (Hassenteufel and de Maillard, 2013).
A second approach to the causal role of ideas is to consider that those that come
from ‘on high’ have to confront the practices of actors who actually implement
policy on the ground. Implementation practices are thus seen as accompanied by
their own sets of social representations which participate strongly in translating, and
sometimes even countering, those which actors with the most formal authority
(for example, ministers) attempt to impose upon them. More precisely, several
66
studies have targeted new kinds of policies which do not set up objectives and
allocate resources in a top down manner, but instead establish rules for inter-
institutional negotiation around locally determined issues (see especially Duran
and Thoenig, 1996; Gaudin, 2007). Meanwhile, some go a stage further to deny
the causal influence of ideas upon policies and their implementation. Envisaging
public action as intrinsically incoherent, ‘big ideas’ such as ‘partnership’ or ‘local
development’ are instead seen as myths which do little to counter the orientations
adopted by materially powerful actors (Desage and Godard, 2005).
Finally, an approach centred upon policy instruments has sought to conciliate an
emphasis upon ideas with another on practices (Halpern et al, 2014; in English,
Lascoumes and Le Galès, 2007). This has been done by underlining the importance
within policy-making of choices made over tools for evaluation, indicators or
incentives, for example, such authors have striven to pin down the influence of
ideas within daily policy practice, rather than simply upon it.
In short, as in other national communities of policy specialists, since the early
1990s the role of ideas has become an omnipresent part of French debates.
However, because positivism is so weak in French social science and constructivism
so strong, in France this debate has not been about whether ideas matter, but
how they do. Together with a focus on institutionalising policy studies at home,
as well as the self-perceived poor standards of English that prevailed at that time,
this emphasis upon how ideas matter contributed strongly to French policy studies
isolating itself from international debates throughout the 1990s.
Institutions and actors within policies

Indeed, it is this question of how ideas matter within policy-making which
eventually, over the course of the 2000s, led a large number of French policy
studies specialists to engage in the new-institutionalism that had influenced and
inspired so many of their colleagues in sociology and political science since
the mid-1980s (Steinmo et al, 1992; Hall and Taylor, 1996). In many cases,
historical institutionalism was imported unquestioningly as an analytical means of
combining a focus on ‘ideas’ with another on stabilised rules and norms, but also
as a vector for aligning French research with that carried out elsewhere (Palier
and Surel, 2005; Palier, 2010). Meanwhile, other research has engaged more
critically with this historical institutionalism by maintaining a more sociological
approach to the actors who change or reproduce institutions (Jullien and Smith,
2012). Overall, if the importance of institutions within policy-making and the
power of institutionalist analysis have ultimately come to be accepted virtually
unanimously in France, nonetheless considerable debate remains over their
relationship with actors.
67
The (late) introduction of institution-inspired approaches
In French social science, institutions – that is, stabilised rules, norms and
expectations – have always been considered crucial. Consequently, many
researchers in this country see much to agree with, but little that is new, in
historical or sociological institutionalism. This is because behaviourism never
took strong root in France where, instead, governmental rules, norms and actors
remained at the core of most early French approaches to public policy (either in
administrative sciences influenced by public law, or even within the sociology
of organisations which has long highlighted the role of governmental rules and
norms). This said, many French specialists of public policies have sought to
contribute to the new institutionalist agenda by centring their work upon the
question of when and how policies institutionalise (Lascoumes and Le Galès, 2011).
A first area of research invested in from this angle has been the economy. Using
the concept of institutionalisation, certain researchers have sought to relaunch
political economy by targeting the ‘political work’ conducted to stabilise or disrupt
markets and the industries within which they operate. The premise here is that
both markets and industries are structured by institutions defined, sociologically,
as stabilised sets of rules, norms and conventions. Although these institutions
are relatively stable, they are never totally at a point of equilibrium because
they are constantly being worked upon politically. Conceptualised as differing
combinations of problematisation (defining social situations as public problems),
instrumentation (setting policy instruments) and legitimation (through discourse
and other symbolic action), this work is political because it either mobilises values
to change or reproduce institutions or, on the contrary, suppresses value-based
action through technicisation (Jullien and Smith, 2014).
The analysts of policy instruments mentioned above have also focused upon
how their objects of study become, or fail to become, institutions in their
own right. Indeed, such work sometimes dovetails with a separate stream of
French research upon ‘knowledge in government’ which focuses upon how
‘cognitive models’ shape and stabilise practices within a range of national and
local administrations (Bongrand et al, 2012). In line with dominant strands of
sociological institutionalism, this approach questions how some frontiers (such
as the one between political and administrative fields) have been institutionalised
through the mobilisation of knowledge.
Third, institutionalisation has also been used as a concept by French scholars
seeking to understand European integration in general, and the autonomisation of
certain European Union organisations in particular (for example, the Commission
or the Court of Justice). Here emphasis placed upon the institutionalisation of
sectors at the European scale has been particularly salutary for studying objects
ranging from industries (Jullien and Smith, 2014) to public health (Guigner, 2012).
Specifically, by focusing upon how and why sectors or policies institutionalise
at the EU scale, this research has shown why improbable ‘candidates’ for EU
government such as health contain increasing levels of EU government, but also
68
why ‘usual suspects’ such as agriculture have actually experienced significant

deinstitutionalisation at the EU scale in recent years.
Neo-institutionalist approaches have been mobilised even more specifically to
produce knowledge about the extent and scope of policy changes. Especially with
regards to social policies, scholars have critically used neo-institutionalist concepts
and studies to analyse the mechanisms of policy change (in particular those that
underline the cumulative effect of successive small reforms: Streeck and Thelen,
2005). In so doing, they have highlighted various constraints on policy changes
(notably due to path dependency processes), but also the reform trajectory of
the French welfare state and, more generally, the politics of welfare reform in
continental Europe (Palier, 2010). The overall claim made here is that changes
of continental European welfare systems must not be analysed in isolation, but as
a succession of reforms that have each opened up new opportunities for change.
Linking institutions and actors
Overall then, since the early 2000s, compatible neo-institutionalisms have certainly
become part of the standard toolbox of French policy specialists. Nevertheless,
both historical and sociological institutionalism continue to cause frictions within
the French field of policy studies because many scholars in this country consider
that these approaches do not take the role played by actors sufficiently into account.
Indeed, it is not by chance that policy studies specialists in France tend strongly
to see themselves as conducting ‘sociologies of public action’. Indeed, to quote
one of its leading figures, Patrick Hassenteufel, the dominant definition of policy
studies in France is one that considers that
the collective construction of public action is the product of

interactions that range from strongly to weakly structured, stable and
coherent; superimpose several levels of government that are more
or less institutionalised; needs grasping on the basis of the resources,
representations and interests of the different actors involved, that is,
by analysing their internal structuration through taking into account
individual actors. (2011, 155)
See also Lascoumes and Le Galès, 2011, 14–16, or Nay and Smith, 2002. Several
examples of empirical research reflecting this perspective help clarify the theory-
driven aim of this definition.
The first concerns EU policy-making where French authors have consistently
pushed for actors to be better taken into account (Saurugger, 2015). In studying
issues such as ‘the usages of Europe’ (Jacquot and Woll, 2004) or how local politics
affects the EU (Pasquier and Weisbein, 2004), the focus has been less upon the
evolving rules and norms so dear to historical institutionalism (Streeck and
Thelen, 2005) and more upon ‘mechanisms of appropriation and socialisation,
69
especially those involving state elites – both administrative and political’ (Radaelli,
2015, 228).
A second example of actor-centred approaches to the institutions of policy-
making concerns research on policy transfer across national borders. More than
neo-institutionalists, French work in this area has emphasised how ‘active national
and local “receptors”’ play a key role in ‘translating’ practices identified abroad into
policy recipes at home (Hassenteufel and de Maillard, 2013, 390). Specifically, how
power is configured within bureaucratic spaces is seen as playing a decisive role.
The central claim made here is that the translation of international norms occurs
within domestic contexts, marked by institutional rivalries, ideational frameworks
and political constraints. From this point of view, processes of ‘convergent
divergence’ and the mixing of diverse components generally predominate.
Ideas emanating from international organisations circulate through the active
interventions of policy intermediaries who redefine these norms and acclimatise
them to local contexts (Clavier, 2013). One example of setting up research
around this claim is to discover how neoliberal ideas (especially those coming
from the Chicago school) have not been translated in a straightforward manner
into national standards and norms, even within supposedly internationalised
competition policies (Smith, 2013). Another example is how the transposition
of ‘new public management’ in Russia has been marked by the legacy of the
past, especially the enduring relations between public sector actors within the
Federal state (Sigman, 2013).
A third approach to actors in policy-making illustrates a combination of
institutionally derived and structuralist analyses when focusing upon the work of
street level bureaucrats within public administrations (Dubois, 2012). According
to this Bourdieu-influenced approach, these agents oscillate between various
identities that are administrative (relying on the strict enforcement of rules) and
personal (involving forms of compassion towards users). These agents interpret
loose administrative categories in ways that depend upon their own trajectories
and social distance towards policy beneficiaries. This ‘critical’ research to policy
studies also highlights the effects upon social services of macro-level changes
towards neoliberal policies.
Finally, socio-historical research on institutions conducted in France also
highlights the role of social and political actors. If this work also underlines the
structuring effects of institutions, they above all seek to identify the contingent
moments before such rules, norms and conventions stabilised. Consequently, the
long-term effects of institutions are studied less, and the microsocial interplay
of actors more, than in similar studies undertaken in other countries (Payre and
Pollet, 2013).
In summary, ever since the 1990s this type of micro perspective on actors has
frequently been adopted by policy studies specialists in France order to grasp who
changes or reproduces institutions, as well as how they do so. If, during the 2000s,
a willingness to import ‘off the shelf ’ theories and concepts began to challenge
this inward-looking way of conducting research and engaging in international
70
scholarship, during that decade it prompted little innovation and considerable new
friction within the domestic field of policy studies itself. Revealingly, it is only
over the past five or six years that both these characteristics have begun to change.
Does policy analysis only study the making of policies, or the state
and politics?
Indeed, revitalising French policy studies has largely come about through the
resuscitation of another often cited ambition of French policy studies: generating
knowledge on wider phenomena, namely the state and politics. This has been
an enduring line of questioning within the domestic field that partly reflects a
disciplinary divide. If sociologists are prone to analyse public policies as incoherent
sets of programmes implemented by a varying range of public and private actors
(thus giving no primacy to political regulation), political scientists have tended
to link the analysis of the multiplication of actors to more macro approaches to
the state and political power (see Musselin, 2005). Indeed, this linkage between
public policies and political phenomenon is a constant source of debate. A fear
often expressed in France is that specialists of policy analysis on the one hand
and those of the state on the other, have created sub-disciplines for themselves
which either fail to communicate, thereby creating a logic of decreasing returns
(Bezes and Pierru, 2012), or forget to really study the linkage between politics,
policies and polity. Here many non-specialists of policy studies blame ‘Anglo-
Saxon’ policy analysis and its plethora of concepts (‘windows of opportunity’,
governance, policy networks, advocacy coalitions, and so on). If this claim is not
completely without foundation, it is more accurate and interesting to highlight
how and why two recent series of publications by French policy specialists have
nevertheless sought to tackle first the transformation of the state and, second,
the links between public policies and political phenomena. Crucially, both these
strands of work have sought to better integrate insights from French and non-
French scholarship, and this while engaging more assertively in international
debates.
Between public policies, elites and the changing state
Indeed, scholars from within policy studies have worked to rebind it into what is
known in France as the sociology of the state, and this by reinvesting in reflection
over elites. In so doing, they are partly reviving a strand of analysis developed by
the original importers of PPA into France who all highlighted the influence of
state elites upon French policies, the importance of civil servants being trained
within the Grand Corps and, more generally, state interventionism in the economy
and society (Thoenig, 1987). But this emphasis upon elites also stems from an
attempt to update knowledge about the state combining tools from policy studies
and administrative sociology.
71
The most formalised approach here is that of ‘programmatic elites’ developed

by Genieys and Hassenteufel (2012; Genieys, 2011). Beginning with work on
what they call ‘the welfare elite’, these authors mapped the civil servants who had
worked for or against the liberalisation of social policies, and health insurance in
particular, within the French state during the 1990s. Biographical statistics were
compiled and a typology of career trajectories was established. Meanwhile, a
wide range of interviews were conducted with these actors in order to grasp the
cognitive, symbolic and material resources or representations they had mobilised
in order to promote or resist policy change. By combining these sociographic
and interpretative methods, they then proposed an analytical framework for
studying trends within states which places competing ‘programmatic actors’ at its
centre. In claiming that this framework is generalisable to other sectors and issue
areas, Genieys and Hassenteufel have thereby put forward a means of comparing
change in governmental activity through studying closely both who governs, and
in the name of what. More precisely still, they have shown, including within
international journals (2015), how these two questions can be rigorously tackled
within empirical research projects. Interestingly, some authors have gone beyond
the sole focus on administrative elites (and, conversely, street-level bureaucrats),
to promote the careful analysis of the work of middle managers within public
administrations in charge of translating general objectives into practice (Barrier
et al, 2015). From a different angle, French analysts of policy instruments have
also sought to reinvest in studying states, here by combining analyses in terms of
instrument type, its relationship to politics and its primary mode of legitimacy
(Lascoumes and Le Galès, 2007, 351; Halpern et al, 2014, 18). Inspired by Michel
Foucault, the aim here has been to track the evolving role of particular instruments
as a means of understanding contemporary government and its legitimacy. Starting
from another point (but with a shared Foucaldian inspiration), some scholars
have explored the politics of administrative reform within the French state and
the diffusion of neo-managerialist recipes and standards (Bezes, 2009), but also
the transformation of the boundaries of public administration since the 1980,
especially through agencification and organisational mergers (Bezes and Le Lidec,
2016). In his latest work, Pierre Muller (2015) joins such authors in considering
that there is a distinct co-relation between the types of policy instrument mobilised
in a polity and its type of state.
Public policy and politics
Debates over policy studies’ treatment of the state are mirrored by another
concerning the way researchers using this perspective study politics. Over the
last decade, three ways of tackling this question have emerged.
The first concerns the role of politicians in public policy-making. Here studies
have increasingly targeted the capacity of actors such as mayors (Douillet and
Robert, 2007) or even the French President (de Maillard and Surel, 2012) to
participate in the production and implementation of public policies. Are they
72
able to build and maintain relations with experts or holders of other types of
legitimacy, and this by building durable teams of advisors and wider networks?
Can they use the specificity of their relationship to the public in order to increase
their influence of policy-making? Has the changing sociological profile of
politicians (for example, higher education levels) had an impact upon how they
work with administrators (Gaudin, 2007)? In terms of theoretical propositions,
however, initial work on ‘political leadership’ (Smith and Sorbets, 2003) still largely
remains to be followed up with systematic and original research. In particular,
the collective dimension of political leadership (for example, a politician’s team
and networks), as well as their communication practices, has yet to give rise to
the generation of quantitative and qualitative data.
Another classical approach to the role of politics in public action is to study
the impact of electoral cycles. Beyond the traditional issue of the electoral cycle
and the question of the effect of political turnover, some authors have questioned
how the political rhythm (as regards the calendar fixed by elections) shapes
policy orientations and calculus (de Maillard, 2006). Others have underlined the
discrepancy between political time and managerial time, especially by focusing
on the actual length in office of ministers and showing that the ministers who
initiate a reform are rarely those who implement it, thereby increasing the effects
of political announcements not followed by action (Siné, 2005). Meanwhile,
other researchers have revived interest in the effect of political parties on public
decisions, by specifying under what conditions governing parties influence public
policies. From this angle, the now longstanding Comparative Agendas Project has
had considerable impact upon research undertaken at Sciences Po Paris, Bordeaux
and Grenoble, in particular (Persico et al, 2012). Some have also used public
policy concepts and tools to study the electoral field (Erhard, 2016).
Finally, and more recently, scholars in France have begun to reinvest in the
question of how public action produces public and political order (Zittoun,
2013). More precisely, here it is claimed that public policies are the mechanisms
through which the disorder of both socio-economic and political life is given
a certain pattern. From this perspective, governing means producing discourses
that shape the way we interpret politics and thereby contribute to the production
of a certain social order. For example, within urban studies, some studies have
underlined how governing the city means orienting behaviours, imposing choices
and controlling territories (Aguilera, 2012).
Conclusion
Much more could and should be said about the development of policy studies
in France over the last two decades. Nonetheless, two points that arise from this
chapter both synthesise the distance it has travelled so far and provide pointers
for the road ahead.
First, much has changed in this national field over the last 30 years, in particular
the insertion of French scholars in international research programmes, together
73
with their usage of concepts and paradigms developed elsewhere. This insertion
has often caused frictions within and without France. This is largely because there
has also been considerable continuity within the field, driven in particular by
the importance of political science within it and the close relationship between
this discipline and French sociology. However, a strong trend within recent
publications suggests that sterile conflicts are currently being replaced by a more
innovative approach to both theory development and engagement in international
research and debates.
Second, and finally, another singularity of French policy studies must also be
underlined: the academic developments traced above have been led by analytical
and theoretical debates (analysis of public policy) that have generally been far
removed from more normative lines of questioning (analysis for public policy).
Indeed, some would argue that in France the academic development of policy
analysis has been conducted almost exclusively from the former perspective, and
this at the expense of providing a form of policy analysis that is directly useful to
decision makers. While refuting this allegation (by underlining the contribution
of policy studies to contemporary French expertise and to the teaching of certain
elites), it is also important to recall that such a trajectory is not specific to France.
As elsewhere, this trajectory has been shaped by the various intellectual and
institutionalised components of French policies that will be explored in other
chapters of the book (notably the isolation of the training schools of top civil
servants from the Academy and the relative undervaluing of policy-oriented work
among academic circles). Nevertheless, as this chapter has highlighted throughout,
over the last 30 years, indirectly but surely, French policy studies has made great
strides in affecting both scholarship and public life.
References
Aguilera, T. (2012) ‘Gouverner les illégalismes. Les politiques urbaines face aux
squats à Paris’, Gouvernement et action publique 1(3), 101–24.
Barrier, J., Pillon, J.-M., Quéré, O. (eds) (2015) ‘Les cadres intermédiaires de la
fonction publique’, Gouvernement et action publique 4(4), 9–171.
Bezes, P., Pierru, F. (2012) ‘État, administration et politiques publiques: les
déliaisons dangereuses’, Gouvernement et action publique 1(2), 41–87.
Bezes, P., Le Lidec, P. (eds) (2016) ‘Politiques de l’organisation’, Revue française
de science politique 66(3–4), 407–33.
Bongrand, P., Gervais, J., Payre, R. (eds) (2012) ‘Les savoirs de gouvernement à
la frontier entre “administration” et “politique”’, Gouvernement et action publique
1(4), 9–20.
Boussaguet, L. (2008) La pédophilie, problème public: France, Belgique, Angleterre,
Paris: Dalloz.
74
Boussaguet, L., Surel, Y. (2015) ‘Des politiques publiques “à la française”?’, in

politiques publiques?, pp. 153–84, Paris: Presses de Sciences Po.
Clavier, C. (2013) ‘Les causes locales de la convergence: La reception des transferts
transnationaux en santé publique’, Gouvernement et action publique 2(3), 395–413.
Crespy, A., Ravinet, P. (2014) ‘Les avatars du néolibéralisme dans la fabrique des
politiques européennes’, Gouvernement et action publique 3(2), 9–29.
Desage, F., Godard, J. (2005) ‘Désenchantement idéologique et réenchantement
mythique des politiques locales’, Revue française de science politique 55(4), 633–61.
Douillet, A.-C., Robert, C. (2007) ‘Les élus dans la fabrique de l’action publique
locale’, Sciences de la société 71, 3–26.
Dubois, V. (2012) ‘Ethnographier l’action publique: Les transformations de l’État
social au prisme de l’enquête de terrain’, Gouvernement et action publique 1(1),
83–101.
Duran, P., Thoenig, J.-C. (1996) ‘L’État et la gestion publique territoriale’, Revue
française de science politique 46(4), 580–623.
Durnova, A., Zittoun P. (2013) ‘Les approches discursives des politiques
publiques’, Revue française de science politique 63(3), 569–77.
de Sciences Po.
Erhard, T. (2016) ‘Penser les politiques électorales’, Gouvernement et action publique
5(1), 9–33.
Faure, A., Pollet, G., Warin, P. (eds) (1995) La construction du sens dans les politiques
publiques, pp. 13-24, Paris: L’Harmattan.
Fouilleux, E. (2003) La politique agricole commune et ses réformes, Paris: L’Harmattan.
Gaudin, J.-P. (2007) Gouverner par contrat: L’action publique en question, Paris: Presses
de Sciences Po.
Genieys, W. (2011) Sociologie politique des élites, Paris: Armand Colin.
Genieys, W., Hassenteufel, P. (2012) ‘Qui gouverne les politiques publiques?
Par-delà la sociologie des élites’, Gouvernement et action publique 1(2), 89–115.
Genieys, W., Hassenteufel, P. (2015) ‘The shaping of the new state elites:
Healthcare policymaking in France since 1981’, Comparative Politics 47(3), 280–95.
Guigner S. (2012) ‘Pour un usage heuristique du néo-institutionnalisme:
Application à la “directive temps de travail”’, Gouvernement et action publique
1(4), 7–29.
Hall, P. (2015) ‘An American perspective to French approaches to public policy-
making’, in L. Boussaguet, S. Jacquot, P. Ravinet (eds) (2015) Une ‘French touch’
dans l’analyse des politiques publiques?, pp. 237–42, Paris: Presses de Sciences Po.
Hall, P., Taylor, R. (1996) ‘Political science and the three new institutionalisms’,
Political Studies 44(5), 936–57.
75
Halpern, C., Lascoumes, P., Le Galès, P. (eds) (2014) L’instrumentation de l’action

publique, Paris: Presses de Sciences Po.
Hassenteufel, P. (2011) Sociologie politique de l’action publique, Paris: Armand Colin,
2nd edn.
Hassenteufel, P., Maillard, J. de (2013) ‘Convergence, transferts et traduction:
Les apports de la comparaison transnationale’, Gouvernement et action publique
2(3), 379–93.
Jacquot, S., Woll, C. (eds) (2004) Les usages de l’Europe, Paris: l’Harmattan.
Jobert, B. (1995) ‘Rhétorique politique, controverses scientifiques et construction
de nouvelles normes institutionnelles: esquisse d’un parcours de recherché’,
in A. Faire, G. Pollet, Ph. Warin (eds) La construction du sens dans les politiques
publiques, pp. 13-24, Paris: L’Harmattan.
Jobert, B., Muller, P. (1987) L’Etat en action, Paris: Presses Universitaires de France.
Jullien, B., Smith, A. (2012) ‘Le gouvernement d’une industrie: vers une économie
politique institutionnaliste renouvelée’, Gouvernement et action publique 1(1):
107–28.
Jullien, B., Smith, A. (eds) (2014) The EU’s Government of Industries, London:
Routledge.
Lascoumes, P. (1995) ‘Les arbitrages publics des intérêts légitimes en matière
d’environnement’, Revue française de science politique 45(3), 396–419.
Lascoumes, P., Le Galès, P. (eds) (2007) ‘Understanding public policy through
its instruments’, Governance 20(1), 1-144.
Colin.
Maillard, J. de (2006) ‘La conduite des politiques publiques à l’épreuve des
temporalités électorales: Quelques hypothèses exploratoires’, Pôle Sud 25, 39–53.
Maillard, J. de, Surel, Y. (eds) (2012) Politiques publiques sous Sakozy, Paris: Presses
de Sciences Po.
Mazur, A., Revillard, A. (2016) ‘Gender policy studies: Distinct, but making the
comparative connection’, in R. Elgie, E. Grossman, A. Mazur (eds) The Oxford
Handbook of French politics, pp. 556–82, Oxford: Oxford University Press.
Muller, P. (2015) ‘Une théorie des cycles d’action publique pour penser les
changements systémiques’, in L. Boussaguet, S. Jacquot, P. Ravinet (eds) Une
de Sciences Po.
Muller, P., Surel, Y. (eds) (2000) ‘Les approches cognitives des politiques publiques:
présentation’, Revue française de science politique, 50(2), 187–350.
Musselin, C. (2005) ‘Sociologie de l’action organisée et analyse des politiques
publiques: deux approches pour un même objet?’, Revue française de science
politique 55(1), 51–71.
Nay, O., Smith, A. (2002) ‘Les intermédiaires en politique : médiation et jeux
d’institutions’, in O. Nay, A. Smith (eds) Le gouvernement du compromis, pp.
149–75, Paris: Economica.
76
Palier, B. (2003) ‘Gouverner le changement des politiques de protection sociale’, in

P. Favre, J. Hayward, Y. Schemeil (eds) Etre gouverné, Paris: Presses de Sciences Po.
Palier B. (ed) (2010) A long good bye to Bismarck?, Amsterdam: Amsterdam
University Press.
Palier, B., Surel, Y. (2000) ‘Les “trois I” et l’analyse de l’Etat en action’, Revue
Pasquier, R., Weisbein, J. (eds) (2004) ‘L’Europe au microscope du local’, Politique
européenne, 12, 5–121.
Payre, R., Pollet, G. (2013) Socio-histoire de l’action publique, Paris: La Découverte.
Persico, S., Froio, C., Guinaudeau, I. (2012) ‘Action publique et partis politiques:
L’analyse de l’agenda legislative français entre 1981 et 2009’, Gouvernement et
Quantin, P., Smith, A. (2013) ‘Risse’s deliberative logic and governance: a critical
engagement’, Critical policy studies 7(3), 263–72.
Radaelli, C. (2015) ‘Un certain regard: “French” public policy analysis in
Europeanization research’, in L. Boussaguet, S. Jacquot, P. Ravinet (eds) Une
de Sciences Po.
Saurugger, S. (2015) ‘The Europeanization of public policy in France: Actor-
centred approaches’, in R. Elgie, E. Grossman, A. Mazur (eds) Oxford Handbook
of French politics, Oxford: Oxford University Press.
Sigman, C. (2013) ‘Le “nouveau management public” en Russie’, Gouvernement
et action publique 2(3), 441–60.
Siné, A. (2005) ‘Politique ou management public: le temps de la politique et
le temps de la gestion publique’, Politiques et management public 23(3), 19–40.
Smith, A. (1999) ‘Developments in French Policy Analysis’, Public Administration
77(1), 111–31.
Smith, A. (2013) ‘Transferts institutionnels et politiques de concurrence’,
Gouvernement et action publique 2(3), 415–40.
Smith, A., Sorbets, C. (eds) (2003) Le leadership politique et le territoire, Rennes:
Presses universitaires de Rennes.
Steinmo, S., Thelen, K., Longstreth, B. (eds) (1992) Structuring politics: Historical
institutionalism in comparative analysis, Cambridge: Cambridge University Press.
Streeck, W., Thelen, K. (eds) (2005) Beyond continuity: Institutional change in
advanced political economies, Oxford: Oxford University Press.
de France.
Thoenig, J.-C. (1987) L’ère des technocrats (2nd edn), Paris: L’Harmattan.
Zittoun, P. (2013) La fabrique politique des politiques publiques, Paris: Presses de
Sciences Po.
77
FIVE
Methods of French policy studies

Claire Dupuy and Philippe Zittoun
Introduction
Methodological debate is central to characterising the scientificity of social

sciences research (Merton, 1957) and policy studies in particular (Lerner and
Lasswell, 1951). This chapter focuses on French policy studies and the methods of
policy analysis. Its objective is to highlight the main features of these approaches
as compared to other national community practices. This task seems somewhat
daunting as explicit discussions on methods among policy specialists are scarce.
This is evidenced by the fact that over the last 30 years, none of the main French
public policy analysis textbooks contain a complete chapter on methodology
(Thoenig, 1985; Mény and Thoenig, 1989; Muller and Surel, 1998; Duran,
1999; Muller, 2015; Gaudin, 2004; Boussaguet et al, 2014; Lascoumes and Le
Galès, 2006; de Maillard and Kübler, 2015; Massardier, 2008; Hassenteufel,
2008). The same can be said for classic French policy studies (Padioleau, 1982;
Dupuy and Thoenig, 1985; Jobert and Muller, 1987; Lascoumes and Le Galès,
2004; Collectif, Laborier and Trom, 2003; Payre and Pollet, 2005; Zittoun,
2013). Recently though, special issues have dealt with methodological issues
(see Gouvernement et action publique, 2012, 1–2 ; Muller et al, 2005). This relative
neglect of methodology is not specific to France; in other national settings as well,
methodological debates rank second in collective discussions. The French case,
however, may be explained by the peculiarity of French policy studies, that is,
the influence of sociology. This has led to a strong reliance on the methods and
debates privileged in sociology studies and thus to the absence of autonomous
methodological debates. This contrasts sharply with methods used in policy
analysis, even though our knowledge on practitioners is still scarce.
This chapter primarily focuses on political science, where in academia in
particular, a clear emphasis has been placed on producing empirical knowledge
on the policy process that is axiologically neutral, as opposed to normative
contributions. There is also a strong tradition of qualitative empirical inquiry in
French studies but relatively few quantitative studies (Lewis-Beck and Bélanger,
2015); the sociological imprint of inquiry tools is obvious, notably semi-structured
interviews, mid-term analyses of the policy process and small-N studies; last, when
it comes to theoretical approaches, emphasis is placed on ideational processes
and, in parallel, rational choice frameworks are rejected. By contrast, there is
greater heterogeneity in terms of methods when the analysis is undertaken by
79
practitioners. Moreover, this type of policy analysis has hardly been formalised and
prioritises the oral transmission of know-how among professional communities.
During the 1970s, however, specially trained engineers (Grand Corps) who then
dominated departmental bureaucracies attempted to develop common methods of
policy analysis. These attempts failed in the 1980s and 1990s because of changes
in power relations between ministries in the aftermath of the economic crisis.
Their failure was also driven by the emergence of alternative expertise within the
public sphere, originating in particular from the decentralisation process of the
then unitary state and a contestation of the central state’s monopolistic expertise.
This chapter therefore suggests that policy studies and the methods of policy
analysis have largely developed in parallel with virtually no intersection. It
illustrates the main features of the methods used by academics and practitioners.
The first section discusses the empirical focus and the methods used in French
academic research to observe the policy process. The second section centres on
the methods that explain this process. The third section examines the methods
used by practitioners for policy analysis.
The empirical observation of the policy process

This section focuses on three issues: the widely debated empirical approaches used
by academics in French policy studies, the issue of time and timeframes when
observing the policy process and, last, data collection methods. As the previous
chapters in this book have established, the emergence of French policy studies as
a subfield within the field of political sciences in the 1980s was linked to the new
representation of the state driven by policy researchers. Policy studies distanced
themselves from administrative and law studies which proposed a coherent and
reified conception of the state. They privileged an empirical vision of the state
through the sociological observation of actors ‘in action’, that is, concretely
making or implementing public policy. The titles of the initial major French
policy studies such as the Concrete State (‘l’Etat au concret’) or the State in Action
(‘L’Etat en action’) illustrate this empirical orientation which sought to ‘open the
black box’ of state policy-making (Jobert and Muller, 1987; Padioleau, 1982).
A strong empirical orientation
Compared to Anglo-Saxon countries, empirical sociology and its methodologies

play an important role in French political science in general (Favre, 1989; 1981)
and in public policy studies in particular. Distinguished from law and administrative
studies from which they originally emerged, the sociological methods used
to observe actors are perceived as key to the scientific analysis. The empirical
focus emerged in the 1970s when policy studies began and became increasingly
important in the 1980s and 1990s.
The previous chapters have shown the importance of the sociology of
organisations as a new theoretical perspective behind the emergence of French
80
policy studies. Rejecting all reification of the state, Crozier applied sociological
methods to empirically observe bureaucracy and paved the way for a new
perspective on state studies in France (Crozier and Friedberg, 1977). He argued
that the empirical observation of actors is the first step to understanding how
an organisation really works and how actors with different interests succeed, or
not, in cooperating and solving major problems in spite of existing conflicts.
Seminal policy studies drew on this method. In his primary work, Jean-Claude
Thoenig, a policy scholar who played a key role in introducing policy studies in
France in the 1970s, combined several sociological methods in the 1970s to study
the role played by high-ranking civil servants in the policy process – the Grand
Corps des Ponts et Chaussées – in the case of the Ministry of Road, Transport and
Housing Policy (Thoenig, 1976). He conducted semi-structured interviews to
grasp their profile, career path and role in the establishment of the new Ministry
of ‘Equipment’. He also conducted a survey on more than 800 mid-level civil
servants and their career trajectories. Bruno Jobert’s studies on the ‘Plan’ (1981)
also resulted from a considerable empirical inquiry into the key civil servants
involved in this planning administration department and drew from a considerable
amount of documentation (Jobert, 1981). The emergence of a policy field in
the 1980s led scholars to develop a policy studies foundation based on empirical
inquiry in the sociological tradition (Zittoun and Demongeot, 2010; Le Galès,
2011; Commaille et al, 2010; Jamous et al, 1969; Lascoumes, 1994; Jobert and
Muller, 1987). Most of these authors produced empirical sectorial policy studies
in fields such as agriculture or the environment and continued to focus on actors
while paying greater attention to their role in the policy process.
Empirical inquiries gained further significance in the 1990s when policy studies
were incorporated in political science. A new generation of researchers followed
in the footsteps of the pioneers, built on this empirical orientation and developed
an important empirical study in their attempt to more widely open the ‘black
box’ of the policy process (for example, Muller, 1995; Jouve and Négrier, 1998;
Zittoun, 2001a; Borraz, 1998; Warin, 1991; Le Galès, 1990; Setbon, 1993;
Smith, 1996). From the early 2000s, another generation investigated new areas
of policy-making pertaining to gender equality in France and in the European
Union (Jacquot, 2015), including policies promoting work–life balance (Morel,
2007; Jacquot et al, 2011), morality policies (Engeli, 2009; Boussaguet, 2008) or
state feminism (Revillard, 2009). The researchers used traditional sociological
methods to study the main actors, their roles and their strategies, and to link their
observations and the policy process. These generations became the main lecturers
for undergraduate and postgraduate students in public policy; they contributed
to constituting this empirical approach influenced by qualitative sociological
methods and the comprehensive perspective as a norm for policy researchers.
This institutionalisation of French policy studies explains why, compared to
other countries, empirical inquiry in France is more qualitative and constructivist
(Hall, 2015), and focuses on how actors influence policy processes. It is interesting
to note that those policy studies proposing some methodological tools to allow
81
for the empirical observation of actors, their representations and their discourses
have received the greatest attention. These include the studies by Hugh Heclo,
Rod Rhodes, Paul Sabatier, Peter Hall, Frank Baumgartner, Vivien Schmidt,
Bryan Jones and Frank Fischer (Heclo and Wildavsky, 1974; Marsh and Rhodes,
1992; Sabatier and Jenkins-Smith, 1993; Hall, 1993; Jones and Baumgartner,
2005; Fischer, 2003). In this sense, rather than attempting to grasp the internal
logic, which is always context-specific, the main empirical issue revolves around
understanding the political dimension (Hall et al, 2015).
The policy process over time
The choice of the study timeframe is another feature of policy analysis methods.
Far from being neutral, this choice has a direct impact on data collection methods.
Generally speaking, policy ethnography is associated with short timeframes,
interviews are associated with short- and medium-term timeframes, and archives
are generally used to study medium- and long-term timeframes. While French
policy studies tend to be dominated by medium-term empirical studies, the other
timeframes can also be found.
Short-term research projects, that is, from a few months to one or two years,
focus on the everyday behaviour of policy actors. Policy ethnography and
participatory or non-participatory observation allow a very precise observation
of the detailed behaviour of policymakers in response to specific events such as
policy reforms, policy implementation, policy-making, or the mobilisation of
interest groups. They also make it possible to explicitly observe the behaviour
of street-level bureaucrats in their everyday lives, as well as the daily routines of
policy beneficiaries. Semi-structured or life-story interviews may also be used as an
additional method. For instance, Vincent Dubois analysed the day-to-day events
surrounding policy implementation by observing interactions between street-
level bureaucrats and policy beneficiaries for several months. He observed both
routine behaviour and variations and compared these to normative bureaucratic
behaviour (Dubois, 2010a). While French policy studies initially focused more
on single decisions and reforms, studies focusing on specific events alone
eventually became rarer. However, along with a growing interest in ethnographic
methodologies, single-event studies have recently re-emerged. Boy, Brudigou,
Halpern and Lascoumes’s work on Grenelle de l’Environnement illustrates this (Boy
et al, 2012). Observing events where more than 1,500 actors came together to
discuss environmental policy, the four authors studied the sociological position
of the actors and the content of their discourses.
Other policy studies have focused on extended timeframes, that is, often
across several decades. Two distinct approaches predominate, although they are
frequently in conflict (Baudot, 2014): historical neo-institutionalism and the
‘socio-historical’ approach. Both approaches pay little attention to the observation
of specific events and privilege the main policy trajectory itself. They focus on
individual actors’ agency and the manner in which stable configuration networks
82
and organisations shape these actors’ behaviours and decision-making. Bruno

Palier’s work on the French welfare system epitomises the use of historical neo-
institutionalism, tightly associated with process tracing. In the volume he edited
on changes in Bismarckian welfare systems, Palier analysed the French reform
trajectory in the post-war period by distinguishing the main dimensions of welfare
reform trajectories (the problems, the reforms, the politics of the reforms, and the
outcomes and consequences of the reforms) (2010, 34), as well as four sequences
of change. Following the standard historical neo-institutionalist understanding
of the unfolding of temporal processes, the analysis pays attention to timing and
lock-in mechanisms.
Studies within the ‘socio-historical’ approach have also examined wide
timeframes and have been influenced by the French historical tradition (from
École Annales to Gérard Noiriel’s understanding of socio-historical research).
Most research tends to focus on policy genesis and institutionalisation processes;
however, in contrast to historical neo-institutionalism, the ‘socio-historical’
approach is less interested in linking the past to the present. Research within
the socio-historical approach considers time to be a key dimension of the policy
analysis approach and, more specifically, of what the proponents of this approach
refer to as state control (étatisation). This approach firmly rejects the use of macro-
sociological concepts or definitions and analytical models, as these are considered
to disregard empirical cases (Payre and Pollet, 2005). This approach privileges
the study of archives.
The third timeframe, that is, the medium-term timeframe, is probably the most
widely-used timeframe in French policy studies. Periods ranging across one or
two decades are frequently investigated through the sociological study of actors
and of policy dynamics. Semi-structured interviews are the most commonly
used method of data collection. This timeframe enables researchers to distance
themselves from specific events and to determine the empirical connection
between past and present events, making it possible to analyse the dynamics of
the policy process. To a certain extent, the medium-term period may work as a
methodological device to prevent any confusion between practitioners’ expertise
and normative orientation. For instance, most chapters in recent publications,
such as the Politiques publiques series published by Presses de Sciences Po analyse
the policy process on a meso-scale. The use of the meso-scale dates back to the
beginning of French policy studies. The early works of Bruno Jobert, Pierre
Muller, Pierre Lascoumes, Jean-Claude Thoenig, and Jean-Gustave Padioleau
studied policy over a period of one decade (Jobert and Muller, 1987; Thoenig,
1976; Lascoumes, 1994; Padioleau, 1982). While these scholars often focused
on specific policy reform events (agriculture policy reform, social policy reform,
equipment reform, environmental reform, and so on), they extended the period
of observation to understand the policy dynamics at the moment the reform took
place. Put differently, they attempted to embed policy reforms into the larger
process of the formation of policy ideas and representations. The policy instrument
approach also privileges observation across one or two decades. In so doing, it
83
makes it possible to analyse instrument choice within the constraining context

of the introduction of the instrument, its implementation, and the resistance it
may trigger (Lascoumes and Le Galès, 2004). The discursive approach primarily
focuses on ten-year studies; while these are longer than studies conducted over
the short term, they are shorter than many medium-term studies which could
span 15 to 20 years. Focusing on the discourse produced by actors during the
policy process and specifically on discrete interactions within the bureaucratic
system, the interviews conducted aim at reconstituting scenes, even though this
is limited by actors’ limited memory capacities.
The predominance of interviews among data collection methods
Despite the strong empirical orientation of French policy studies from the outset,
it is surprising that data collection methods were not discussed until the 2000s,
with the notable exception of Samy Cohen’s widely referenced edited book on
elite interviews. The recent surge in methodological attention is driven by the
broader context of political science, where methodological issues have made
their way to the top of scholars’ collective agenda. Since the beginning, French
policy researchers have used various methods and there has been considerable
diversification over the last few years.
Among the methods for data collection, however, interviews, and semi-
structured interviews in particular, predominate. In a vivid debate published by
the Revue française de science politique (Bongrand and Laborier, 2005; Pinson and
Sala Pala, 2007), the contributors agreed that semi-structured interviews, along
with documentary analysis, are the most widely used method of data collection
in French policy studies. Bongrand and Laborier have shown that recent PhD
dissertations labelled as ‘policy studies’ largely resort to semi-structured interviews.
Based on French policy studies, we may identify four main uses of interviews.
First, most interviews see the interviewee as a witness of the policy process under
study, and they are intended to collect oral history in order to piece together the
often discontinuous and conflicting phases of the policy-making process. These
interviews focus on identifying the different steps of negotiation, opposition and
consensus-building in the policy arena, as well as the associated justifications,
competing definitions and narratives. Second, interviews are used to analyse
stakeholders’ ideas, representations and interests. In such instances, the policy
process is more likely to be considered continuous and progressive and the research
focus is more synchronic than diachronic. The objective of such interviews is
to discuss with stakeholders their definition of policy, their understanding of its
potential consequences, the problems these policies solve, the public they affect,
the values they mobilise, and so on. Moreover, the researchers try to collect the
representations of the interviewed stakeholders and to identify the subjective
definition of their own interest, as well as their preferences and motivations. Third,
interviews may focus on actors’ trajectories and social characteristics. Following
a sociological perspective on the study of elites (Bourdieu and Christin, 1990;
84
Genieys and Hassenteufel, 2012), these interviews primarily trace actors’ careers
and education in order to assess the professional and social networks to which they
belong and their role in the policy process. A fourth use of interviews involves
actors’ own descriptions and representations of their own practices. Attention is
paid to the capacity of the actors to justify and discuss their actions, as justification
is perceived as playing an important role in explaining preferences.
Surprisingly enough, although (semi-structured) interviews are a dominant
method of French policy studies, few studies have addressed why this specific
data collection method is privileged. Indeed, Bongrand and Laborier (2005)
highlight this issue and emphasise the routinised and non-reflexive use of semi-
structured interviews in policy studies. Methodological discussions mainly focus
on the issues raised by semi-structured interviews (for instance, memory issues,
power relationships between the interviewee and the interviewer, the absence of
reflexivity on the part of interviewees) and attempt to identify how to circumvent
them (for a discussion, see Pinson and Sala Pala, 2007).
A similar absence of collective methodological debate can be observed with
documentary sources, irrespective of whether they are grey literature – including
administrative and expertise reports – administrative documents and archives,
personal notes and e-mails, newspaper articles or websites. Practically all policy
studies rely on such material. Although historians in particular have developed
comprehensive methods to study archives and, more broadly, written texts, there
has been no collective methodological debate (for a tentative analysis, see Dupuy
and Pollard, 2012a).
Other methods of policy studies such as policy ethnography and actors’
prosopography have been used increasingly and have gained prominence over
recent years. Interestingly, these strands of research focus on policy actors, both
empirically and analytically. Analysing policy actors helps explain policy change or
inertia (Dubois, 2012; 2010a; Genieys and Hassenteufel, 2012). Some proponents
of policy ethnography argue that this method is instrumental to uncovering the
‘implementation trick’; this refers to the idea that the contradictions between
policies as enacted and communicated by politicians, and as implemented within
a policy context given the resources available and the street-level bureaucrats
can only be perceived through an ethnographic study of policy implementation
(Dubois, 2012). In this respect, ethnographic implementation studies are perceived
as studies that document policy changes in general and not merely policy
implementation per se. In a similar vein, other proponents of policy ethnography
argue that this method allows policy studies to connect the four dimensions of a
policy, that is, policymakers, beneficiaries, street-level bureaucrats and other policy
brokers (Belorgey, 2012), and thus fully understand a given policy. Other studies
contend that actors’ prosopography coupled with other methods is a privileged
methodological tool to study policy change. In a cross-sectoral comparative study,
Patrick Hassenteufel and William Genieys showed that combining methods from
the sociology of elite – including actors’ sociography, reputation and position –
with a cognitive policy analysis triggered significant advances in the analysis of
85
policy change by endogenising its explanation (2012). They argued that policy
change in healthcare and the military sector might be explained by the ideas and
roles of groups of ‘programmatic elites’ that had gained autonomy from elected
politicians, designed a coherent understanding of the role of the central state in
both policy sectors and subsequently shaped policy changes.
Overall, qualitative methods of empirical investigation are dominant in French
policy studies. It is true for mainstream policy studies as well as for less-developed
empirical and analytical perspectives, like feminist comparative policy studies
(Mazur, 2009). This strand of research relies essentially on qualitative methods of
data collection. Its main difference with mainstream studies is that these works
are less dependent upon a sector-based analysis of policy-making and pay more
attention to its cross-sectoral dimension (Perrier, 2013; Engeli and Perrier, 2015).
While quantitative analyses have also been undertaken (in particular through
the French team of the Comparative Agenda Project [Persico et al, 2012]) they
remain at the margins of French policy studies.
Methodological research design

To discuss the process of explanation building in French policy studies and the
ways in which it relates to methodological choices, this section presents some of
the most influential theories and popular analytical tools as well as comparative
research designs.
Popular theories and analytical tools with which to study the policy process
This chapter does not pretend to propose an exhaustive review of the analytical
tools used in French policy studies. Rather, it focuses on three analytical models
largely used to study the policy process. Each model outlines the scope of
observation, the main issues to solve and identifies the most effective explanation.
The first model is the step-by-step model referred to in France as the ‘Jones
framework’ in reference to the Charles Jones textbook (Jones, 1970). It was
imported by Jean-Claude Thoenig in the early 1980s as an analytical method to
analyse the policy process across different stages (Thoenig, 1985, 18–37). This
model was certainly instrumental in encouraging the shift from the French classical
sociology of bureaucracy to the emerging field of public policy; while the focus
on policy actors remained constant, attention also began to be paid to institutions
and outputs. This analytical method involves empirically isolating each phase and
viewing the link between them as problematic. Following Kingdon’s perspective,
the model separately analyses policy formulation, the problem agenda-setting and
the political activities. It also studies the impact these phases have on each other
during the analysis of policy dynamics, actors and ideas in each phase (Thoenig,
1985; Mény and Thoenig, 1989; Lascoumes and Galès, 2006; de Maillard and
Kübler, 2015; Duran, 1999; Hassenteufel, 2008).
86
Second, the cognitive and normative approach plays a key role in French policy
studies. This analytical framework emerged at the end of the 1980s and was
inspired by Kuhn’s perspective on a scientific paradigm (Kuhn, 1962). Initially
developed by Bruno Jobert and Pierre Muller, this approach was an alternative
to the first analytical model as it specifically focused on crisis and policy change
by paying more attention to how ideas influenced the policy process. While this
model continues to question the link between the configuration of actors and
the policy process, it primarily zeroes in on identifying three stages, that is, policy
stability, policy crisis and policy change (Zittoun, 2009), and on empirically
observing the transformation from crisis to stability.
Third, the pragmatic approach has gained importance in public policy over
the last few years (Durnova and Zittoun, 2013). Influenced by Joseph Gusfield,
by Bruno Latour’s sociology of science, and by Bolstanski’s pragmatic sociology
(Gusfield, 1984; Boltanski, 2009; Latour, 1990), this approach focuses on the
career of instruments or policy proposals rather than on a specific period. It
considers instruments to be dynamic and suggests that analytical methods should
focus on the changes observed with regard to instruments and proposals and take
their failure and disappearance into account.
Comparative research designs
One important feature of the research design of policy studies in France is the
comparative method. Since the early 2000s, comparative analyses have become
increasingly popular in French policy studies (Boussaguet and Dupuy, 2014). This
was a period when French policy studies took stock of their latest developments
and when calls were made to ‘breathe new life’ into policy analysis (Hassenteufel
and Smith, 2002). Specifically, in order to counteract what was conceived of as
policy studies-specific explanations, due in part to the use of jargon, calls were
made to ‘mainstream’ policy studies’ analytical tools into standard political science
(in the case of EU policy, see Hassenteufel and Surel, 2000). In addition, over
the same period, new generations of scholars started to explore then uncharted
territories. Europeanisation of public policy and regionalisation of public policy
students thus initiated a research path distinct from the original works in French
policy studies due to their emphasis on comparative designs (Pasquier, 2004;
Le Galès, 2003; Woll and Jacquot, 2010). A growing interest in comparative
analysis grew out of this context not only within the European Union but also
at the international level (Fouilleux, 2015), in Latin America and in African
countries. Patrick Hassenteufel’s contributions were instrumental in discussing
comparative designs from a methodological point of view and raising awareness
of their pitfalls and advantages (2000; 2005). In particular, he highlighted four
designs that largely resemble comparative designs but are essentially different. For
instance, ‘Canada dry comparisons’ typically propose several single case-studies
but provide no common analytical frameworks or even lines of questioning.
Similarly, ‘Office comparisons’ are exclusively based on documentation available
87
on the web but propose no further investigation. ‘Jivaro comparisons’ refer to

studies based on quantitative indicators that over-simplify the phenomenon
under study by assessing only one aspect of it. Last, ‘ventriloquist comparisons’
are works where case studies are designed in such a way that they must confirm
the initial hypotheses without leaving room for any alternative explanations.
Interestingly, however, this new popularity of the comparative analysis failed to
trigger sustained collective methodological discussions (for one exception, see in
particular Dupuy and Pollard, 2012b, on the issues raised by compound research
designs investigating comparative subnational policies in several national settings).
In this respect, French policy studies are very similar to the field of policy analysis
in general where, until very recently, scholars have mostly relied on discussions
of comparative politics to ground their comparative research designs without
explicitly reflecting on their potential variation when applied to policy studies
(Engeli and Rothmayr Allison, 2014; Levi-Faur, 2006).
Case-oriented comparative research designs distinctly dominate, as opposed
to variable-oriented studies which are rather scarce. This observation is largely
dependent on the fact that French policy studies mainly rely on qualitative research
methods much more than on quantitative data analyses. Small N comparative
studies are most common. Comparative case studies investigate a wide range of
issues, from policy process to agenda-building and problem definition. On top of
international or cross-sectoral comparative designs, some studies have also relied
on ‘transnational comparisons’ that pay attention to interdependences between
the cases at hand (Hassenteufel, 2005). These include policy convergence or
policy transfer studies (Hassenteufel and de Maillard, 2013).Classic and compound
research designs are most common among comparative policy studies in France.
Among common comparative research designs, the comparison of one policy
across several national settings is prominent. While some studies seek to analyse
how governments react when faced with similar problems, others explore processes
of policy change. Cornelia Woll’s (2014) recent study of bank bailouts in the
United States, the United Kingdom, France, Germany, Ireland and Denmark
provides an example of the former. Her book seeks to explain the significant
variation of bank rescue schemes from the onset of the financial crisis in 2007,
in terms of overall amount (share of national income) and constraints put on
participating institutions. This research was designed as a paired comparison of the
most similar cases with regard to the type of market economy, the role of banks
in the financial system, and the tradition and ideology vis-à-vis state control over
banks. Despite the similarities expected to account for similar reactions to the
banking and financial crisis, substantial differences were observed in each paired
comparison bank bailout scheme. Woll argued that the differences were due
to the varying ability of the banking sector to organise itself and to its political
activity, that is, to banks’ collective inaction. Other classic comparative designs
have investigated processes of policy change. Charlotte Halpern and Patrick
Le Galès’ (2011) comparative work on EU environmental and urban policy
instruments is exemplary in this regard. The authors focus on EU policy-making
88
and on the supposedly new instruments of the supposedly new EU governance.

The comparison between EU policy instruments in two policy sectors is thus
instrumental in explaining EU environmental and urban policy change and
institutionalisation by emphasising how the choice and combination of policy
instruments relate to the Commission’s strategy to expand its scope of intervention
as well as the limitations on its resources in doing so.
In addition to these classic uses of comparative designs in French policy studies,
compound research designs are well developed. The fields of territorial and
urban policies have been key contributors to the development of such research
designs. In order to study territorial transformations and the ensuing policy
change, most research designs articulate three dimensions: sub-national policies,
national settings, and time. Gilles Pinson’s (2009) book on urban projects and the
use of this policy instrument by middle-sized cities provides a good illustration.
The book explores the changing role of cities as actors in multilevel systems.
It studies the shift over time from state-led hierarchical planning to city-led
horizontal urban projects development. The empirical data rely on a comparative
approach involving four longitudinal and cross-national case studies of the cities
of Marseille and Nantes, both in France, and Venice and Turin in Italy. Based
on this comparative design, Pinson shows the rise of European cities and their
key role in attracting resources, shaping their own development and, overall,
governing European societies.
Practitioners’ policy analysis methods

As discussed in previous chapters, there is an important gap between the academic
policy community and policy practitioners. While the former focus on the
policy process and, more specifically, on practitioners to understand their role
in the process, they keep the latter at a distance to better study them as objects.
In their quest for practical problem-solving methods, policy practitioners keep
these academic policy studies at distance and turn to practitioners’ know-how.
Consequently, the gap between academics who study the policy process and
practitioners who use policy analysis has led to a lack of analytical models taught
at universities or Grandes Écoles; moreover, these models are not shared with
practitioners. Another consequence is the lack of visibility of the policy analysis
practitioners use. The policy analysis methods used by practitioners are primarily
transmitted orally rather than through textbooks or written forms which are
easier to grasp.
Considering this invisibility, we would like to propose two ways of grasping a
modest part of this fragmented image of the analytical method used. The first one
involves observing the Grand Corps where we find high-ranking civil servants and
a common training and profile. The second one takes on a sectoral perspective
where a predominant method is used in a specific period to observe some of the
main methods used to analyse a policy. By combining these two perspectives,
89
we can identify four examples which highlight the success and failure of some
policy analysis models.
The partial success of the use of mathematical models to analyse policy
Depending on the period and the sector, different mathematical models have
emerged and been developed to analyse policy. This success, which is often short-
lived, can be easily linked to the importance of the Grand Corps of state engineers
who have an expert knowledge of mathematics and who occupy high-ranking
positions within a few important Ministries such as Road, Transport, Energy,
Environment, Housing, and so on. Used as a method, mathematical formalisation
allows them to produce analyses they can use to bolster their dominant position
and their influence.
Two examples clearly demonstrate these engineers’ importance. The first
revolves around the housing policy with the development in the 1970s of the ‘Polo’
model which was the first French version of the US Planning-Programming-
Budgeting System (PPBS) (Zittoun, 2001a). This rationalisation of the budget
choice (Rationalisation des Choix Budgétaires [RCB]) method was imported through
a specific practitioner network, the Grand Corps of statistician public administrators
of the National Institute of Statistics and Economic Studies (administrateur de
l’Institut national de la statistique et des études économiques [INSEE]). These engineers
went on to occupy the key positions at the new Department of Prospective
created in 1965 within the Ministry of Economy and, more specifically, within
the sub-department of RCB (Spenlehauer, 1998). Polo’s promoters primarily
sought to build a mathematical model to simulate human behaviour ‘objectively’
and proposed a method to help decision-makers make the ‘best’ choice through
cost–benefit analyses. This model primarily posits that when choosing where
they will live, people overemphasise the utility function. This makes it possible
to grasp their behaviour and, especially, to make them predictable. Consequently,
engineers can create different fictions in which they can simulate the impact of a
policy proposal and make it possible to compare ‘objectively’ the effect of different
attempts. Considered too costly for the Ministry of Finance, the housing policy
was evaluated using this model. The main instrument which helped finance
social housing was compared to another instrument to directly finance people
living under the minimum wage. Based on Polo’s results, two reports were issued
in 1971 and 1975 which contributed to the housing policy reform approved in
1977. Both reports provoked much debate, controversy and counter-expertise
in the housing sector, even though ultimately, the reform did not follow the
‘Polo’ model recommendations. Considered a housing policy failure, this method
progressively disappeared, much like in the United States.
The second example is drawn from the transport policy. Similar to Harold
Mazoyer’s study, some high-ranking civil servants, that is, engineers in particular,
have developed different mathematical models which allow the simulation of
the behaviour of people based on the assumption that they generally choose the
90
‘fastest’ mode of transport. Developed in the 1960s, this different mathematical

model simulated behaviour and developed a fictional account of the impact of
a policy proposal, making it possible to compare different possible choices. The
transport sector was one of the rare cases where the model was developed not
only by practitioners but also within academia by engineer-economists using
the rational choice theory to disseminate and justify the model. In the transport
policy, the monetisation of behaviour helped produce expertise and knowledge
to analyse policy and inform decision-makers. While this method has been
challenged over the last 20 years and has thus lost influence, a few ‘old generation’
experts continue to use it.
The ‘bricolage’ method, a dominant know-how practice
With the exception of some specific sectors and periods, it is quite difficult to
identify a specific method used to analyse policy. By contrast, bricolage, which can
be illustrated through the following three cases, seems to prevail.
The case of the Economic Policy Turn in 1982–85 reveals how some of the
analytical methods used to study policy can be built pragmatically, how they fail
to have any real links to the academic field, and how they appear as the result
of bricolage. In 1970, a homogenisation process within the Ministry of Finance
resulted in less diversity and in the increased presence of some members of the
Grand Corps such as the tax inspectorate (L’inspection des finances) and INSEE
administrators (Jobert, 1994). In contrast to the Keynesian model for the
analysis of the economic and finance policy, a new analytical model referred to
as the ‘désinflation competitive’ was developed (Zittoun, 2001b). Far from being
a theoretical model built by economists, this model was built by civil servants
lacking theoretical knowledge of economics; indeed, the model’s link to economic
theory is highly elusive (Lordon, 1997). Second, although this model is based
on affirmative statements and causal chains (such as ‘exportation is the only way
to get growth’), there are no academic references to back them up. Third, these
statements provide an analytical framework to distinguish ‘relevant’ policy from
‘irrelevant’ policy. The success of this analytical framework can be explained by
the importance of the network that supported and disseminated it, specifically in
the Ministry of Economy through the Grand Corps of finance inspectors.
The second example reveals that some new institutions responsible for analysing
and evaluating policy never discuss methods. In 2007, a specific institution was
established and charged with the specific task of analysing public policy. Initially
known as the General Review of Public Policies (Révision Générale des Politiques
Publiques [RGPP]), it changed its name to Modernisation of Public Policy
(MAP). The MAP has been very active – between 2012 and 2014, it published
more than 60 policy evaluations. These evaluations were essentially produced by
high-ranking civil servants from the different inspection bodies, Within the SG
MAP, these civil servants developed an ‘official’ method with three components:
a collaborative process, a double focus on content and its social acceptability, and
91
the different timescales. The focus on content generally used the MCA method,
with specific attention being paid to criteria such as innovation, transparency,
efficiency or relevance, and to the development of some technical tools such as
econometric studies, opinion surveys, statistics, impact studies and interviews.
The large number of references to all possible techniques reveals the ‘bricolage’
aspect developed by the method. Given this wide repertoire, the evaluator can
do as he/she wishes and uniformity within the different evaluations proposed can
hardly be identified. It is worth noting that most evaluations essentially involve
the synthesis of all existent information and analyses. It is rare for an evaluation
procedure to integrate a real autonomous inquiry and a specific research aspect
to produce new knowledge.
The third example is the Cour des Comptes, a specific institution responsible for
observing public policy from a financial perspective. This institution comprises
high-ranking civil servants from ENA who have received generalist training. This
strategy is very similar to the one used by SG MAP to evaluate public policy. The
institution has a very wide repertoire of evaluation tools which incorporate the
participation of actors. While no specific knowledge is produced, the institution
synthesises and analyses existing knowledge (for instance, the homeless policy
published in 2012).
Conclusion
Two separate worlds produce policy knowledge. They both fail to systematically
develop any specific literature about how to conduct an inquiry, what kind of
method to choose and how to prevent or to take into account the multiple biases
that result from these choices. This has led to two different situations.
On the one hand, researchers produce knowledge using a relatively homogenous
method focused on qualitative inquiry and inspired by sociology. Policy studies first
focus on the role of actors and the importance of ideas and institutions in the policy
process. Moreover, there is a high and common tendency to use semi-structured
interviews and small N comparative case studies as well as process-tracing.
On the other hand, high-ranking civil servants who are members of the
Grand Corps undertake analyses and summarise already existent data in a very
heterogeneous manner. This methodological heterogeneity, which has encouraged
little debate, reveals that the method of knowledge production is perceived as
insignificant and that the bureaucratic elite position and the practical knowledge
it produces is perceived as central in policy analysis (see the chapter on Civil
Servants).
References
Baudot, P.-Y. (2014). ‘Le temps des instruments: Pour une socio-histoire des
instruments d’action publique’, in C. Halpern, P. Lascoumes, P. Le Galès (eds)
L’instrumentation de l’action publique, pp. 193–236, Paris: Presses de Sciences Po.
92
Belorgey, N. (2012) ‘De l’hopital à l’Etat: Le regard ethnographique au chevet

de l’action publique’, Gouvernement et action publique 2(2), 9–40.
Boltanski, L. (2009) De la critique: Précis de sociologie de l’emancipation, Paris:
Gallimard.
Bongrand, P., Laborier, P. (2005) ‘L’entretien dans l’analyse des politiques
publiques: Un impense methodologique?’, Revue française de science politique
55(1), 73–111.
Borraz, O. (1998) Gouverner une ville: Besançon 1959–1989, Rennes: Presses
universitaires de Rennes.
Boudon, R. (1992) ‘Subjective rationality and the explanation of social behaviour’,
in H.A. Simon, M. Egidi, R. Viale, R. Marris (eds) Economics, Bounded Rationality
and the Cognitive Revolution, pp. 123–47, Aldershot: Edward Elgar.
Bourdieu, P., Christin, R. (1990) ‘La construction du marche: Le champ
administratif et la production de la “politique du logement”’, Actes de la recherche
en sciences sociales 81–2, 65–85.
Paris: Dalloz.
Boussaguet, L., Dupuy, C. (2014) ‘L’analyse des politiques publiques à l’épreuve
de la comparaison’, Revue internationale de politique comparée 21(1), 97–119.
Boussaguet, L., Jacquot, S., Ravinet, P. (2014) Dictionnaire des politiques publiques
(4th edn), Paris: Presses de Sciences Po.
Boy, D., Brugidou, M., Halpern, C., Lascoumes, P. (2012) Le Grenelle de
l’environnement: Acteurs, discours, effets, Paris: Armand Colin.
Commaille, J., Dumoulin, L., Robert, C. (2010) La juridicisation du politique, Paris:
Droit et société classics, Librairie générale de droit et de jurisprudence- Lextenso
éditions, https://halshs.archives-ouvertes.fr/halshs-00458208.
Crozier, M., Friedberg, E. (1977) L’acteur et le système: Les contraintes de l’action
collective, Paris: Seuil.
Dubois, V. (2010a) ‘Politiques au guichet, politique du guichet’, in O. Borraz, V.
Guiraudon (eds) Politiques publiques, 2, pp. 265-286, Paris: Presses de Sciences Po.
Dubois, V. (2010b) La vie au guichet (3rd edn), Paris: Economica.
Dubois, V. (2012) ‘Ethnographier l’action publique: Les transformations de l’Etat
social au prisme de l’enquête de terrain’, Gouvernement et action publique 1(1),
83–101.
Dupuy, C., Pollard, J. (2012a) ‘A quoi servent les rapports administratifs
d’expertise? Retour sur une source écrite centrale en sociologie de l’action
publique’, Political Science Working Paper Series (Travaux de science politique) 53,
Lausanne: Université de Lausanne.
Dupuy, C., Pollard, J. (2012b) ‘La comparaison de politiques publiques
infranationales: Méthodes et pratiques’, Revue internationale de politique comparée
19(2), 7–14.
Dupuy, F., Thoenig, J.-C. (1985) L’administration en miettes, Paris: Fayard.
93
Duran, P. (1999) Penser l’action publique, Paris: Librairie Générale de Droit et de

Jurisprudence (LGDJ)/Montchrestien.
Durnova, A., Zittoun, P. (2013) ‘Les approches discursives des politiques
publiques’, Revue française de science politique 63(3), 569–77.
Engeli, I. (2009) ‘La problématisation de la procréation médicalement assistée en
France et en Suisse’, Revue française de science politique 59(2): 203–19.
de Sciences Po.
Engeli, I., Rothmayr, Allison, C. (eds) (2014) Comparative policy studies: Conceptual
and methodological challenges, Basingstoke: Palgrave Macmillan.
Favre, P. (1981) ‘La science politique en France depuis 1945’, International Political
Science Review 2(1), 95–120.
Favre, P. (1989) Naissance de la science politique en France, 1870–1914, Paris: Fayard.
Fischer, F. (2003) Reframing public policy, New York: Oxford University Press.
Fouilleux, E. (2015) ‘Au-delà des Etats en action…la fabrique des politiques
publiques globales’, in L. Boussaguet, S. Jacquot, P. Ravinet (eds) Une ‘French
touch’ dans l’analyse des politiques publiques?, pp. 287–318, Paris: Presses de
Sciences Po.
Gaudin, J.-P. (2004) L’action publique: Sociologie et politique, Paris: Presses Sciences
Po and Dalloz.
Par-delà la sociologie des élites ?’ Gouvernement et action publique 2(2), 89–115.
Gusfield, J.R. (1984) The culture of public problems: Drinking-driving and the symbolic
order, Chicago, IL: University of Chicago Press.
Hall, P. (1993) ‘Policy paradigms, social learning and the state: The case of
economic policymaking in Britain’, Comparative Politics 25(3), 275–96.
Hall, P., Schmidt, V. and Thatcher M. (2015) ‘Cognitive approaches: A French
touch?’, in L. Boussaguet, S. Jacquot, P. Ravinet (eds) Une ‘French touch’ dans
l’analyse des politiques publiques?, pp. 237–62, Paris: Presses de Sciences Po.
Halpern, C., Le Galès, P. (2011) ‘Pas d’action publique autonome sans instruments
propres. Analyse comparée et longitudinale des politiques environnementales et
urbaines de l’Union Européenne’, Revue française de science politique 61(1), 51–78.
Hassenteufel, P. (2000) ‘Deux ou trois choses que je sais d’elle: Quelques
enseignements tirés d’expériences de comparaison européenne’, in Centre
universitaire de recherches sur l’action publique et le politique (CURAPP)
(eds) Les méthodes au concret, pp. 105–24, Paris: Presses universitaires de France.
Hassenteufel, P. (2005) ‘De la comparaison internationale à la comparaison
transnationale: Les déplacements de la construction d’objets comparatifs en
matière de politiques publiques’, Revue française de science politique 55(1), 113–32.
Hassenteufel, P. (2008) Sociologie politique: l’action publique, Paris: Armand Colin.
94
Hassenteufel, P., Maillard, J. de (2013) ‘Convergence, transferts et traduction.

Les apports de la comparaison transnationale’, Gouvernement et action publique,
3(3), 377–93.
Hassenteufel, P., Smith, A. (2002) ‘Essoufflement ou second souffle? L’analyse
des politiques publiques “à la française”’, Revue française de science politique, 52(1),
53–73.
Hassenteufel, P., Surel, Y. (2000) ‘Des politiques publiques comme les autres?
Construction de l’objet et outils d’analyse des politiques européennes’, Politique
européenne 1(1), 8–24.
Heclo, H., Wildavsky, A. (1974) Private Government of Public Money, Basingstoke:
Macmillan.
Jacquot, S. (2015) Transformations in EU Gender Equality: From Emergence to
Dismantling, Basingstoke: Palgrave Macmillan.
Jacquot, S., Ledoux, C., Palier, B. (2011) ‘The Europeanization of reconciliation
policies in France: Boasting…but learning’, European Journal of Social Security
13(1): 68–87.
Jamous, H., Commaille, J., Pons-Vignon, B. (1969) Sociologie de la décision: La
réforme des études médicales et des structures hospitalières, Paris: Centre national de
la recherche scientifique (CNRS).
Jobert, B. (1981) Le social en plan, Paris: Les éditions ouvrières.
Jobert, B. (1994) Le tournant néo-libéral en Europe, Paris: L’Harmattan.
Jobert, B., Muller, P. (1987) L’Etat en action: Politiques publiques et corporatismes,
Jones, B.D., Baumgartner, F.R. (2005) The Politics of Attention: How Government
Prioritizes Problems, Chicago, IL: University of Chicago Press.
Jones, C.O. (1970) An Introduction to the Study of Public Policy, Belmont, CA:
Wadsworth.
Jouve, B., Négrier, E. (1998) Que gouvernent les régions d’Europe?: Échanges politiques
et mobilisations regionals, Paris: L’Harmattan.
Kuhn, T.S. (1962) ‘The structure of scientific revolutions’, Chicago, IL: University
of Chicago Press.
Laborier, P., Trom, D. (2003) Historicités de l’action publique, Paris: Presses
Lascoumes, P. (1994) L’éco-pouvoir: Environnements et politiques, Paris: La
Découverte.
Lascoumes, P., Le Bourhis, J.-P. (2014) ‘En guise de conclusion. Les résistances aux
instruments de gouvernement: Essai d’inventaire et de typologie des pratiques’, in
C. Halpern, P. Lascoumes, P. Le Galès (eds) L’instrumentation de l’action publique,
pp. 493–520, Paris: Presses de Sciences Po.
Lascoumes, P., Le Galès, P. (2004) Gouverner par les instruments, Paris: Presse de
la Fondation Nationale de Sciences politiques.
Colin.
95
Latour, B. (1990) La science telle qu’elle se fait: Anthropologie de la sociologie des sciences
de langue anglaise, Paris: La Découverte.
Le Galès, P. (1990) Politique urbaine et développement local: Une comparaison franco-
britannique, Paris: L’Harmattan.
Le Galès, P. (2003) Le retour des villes européennes: Sociétés urbaines, mondialisation,
gouvernement et gouvernance, Paris: Presses de Sciences Po.
Le Galès, P. (2011) ‘Public policy research: Deconstructing the French touch’,
Critical Policy Studies 5(2), 201–7, doi:10.1080/19460171.2011.576533.
Lerner, D., Lasswell, H.D. (1951) The Policy Sciences: Recent Developments in Scope
and Method, Stanford, CA: Stanford University Press.
Levi-Faur, D. (2006) ‘Varieties of regulatory capitalism: Getting the most out of
the comparative method’, Governance 19(3), 367–82.
Lewis-Beck, M., Bélanger, E. (2015) ‘Quantitative methods in political science:
Research in France and the United States’, French Politics 13, 175–84.
Lordon, F. (1997) Les quadratures de la politique économique, Paris: Albin Michel.
Maillard, J. de, Kübler, D. (2015) Analyser les politiques publiques (2nd edn),
Fontaine: Presses universitaires de Grenoble.
Marsh, D., Rhodes, R.A.W. (1992) Policy Network in British Government, Oxford:
Clarendon Press.
Massardier, G. (2008) Politiques et actions publiques, Paris: Armand Colin.
Mazur, A. (2009) ‘Les mouvements féministes et l’élaboration des politiques dans
une perspective comparative: Vers une approche genrée de la démocratie’, Revue
de France, 2015.
Merton, R.K. (1957) Social Theory and Social Structure, New York: Simon and
Schuster.
Morel, N. (2007) ‘Le genre des politiques sociales: L’apport théorique des “gender
studies” à l’analyse des politiques sociales’, Sociologie du travail 49, 383–97.
Muller, P. (1995) ‘Les politiques publiques comme construction d’un rapport
au monde’, in A. Faure, G. Polet, P. Warin (eds) La construction du sens dans les
politiques publiques, pp. 13–23. Paris: L’Harmattan.
Muller, P. (2015) Les politiques publiques (11th edn), Paris: Presses universitaires
de France.
Muller, P., Surel, Y. (1998) L’analyse des politiques publiques, Paris: Montchrestien.
Muller, P., Palier, B., Surel, Y. (2005) ‘L’analyse politique de l’action publique:
Confrontation des approches, des concepts et des méthodes’, Revue française de
science politique 55(1), 5–6.
Palier, B. (2010) A Long Goodbye to Bismarck? The Politics of Welfare Reform in
Continental Europe, Amsterdam: Amsterdam University Press.
Pasquier, R. (2004) La capacité politique des régions: Une comparaison France–Espagne,
Rennes: Presses universitaires de Rennes.
96
Payre, R., Pollet, G. (2005) ‘Analyse des politiques publiques et sciences

historiques’, Revue française de science politique 55(1), 133–53.
Perrier, G. (2013) ‘Politiques publiques’, in C. Achin, L. Bereni (eds) Dictionnaire
genre et science politique, pp. 395–407, Paris: Presses de Sciences Po.
Persico, S., Froio, C., Guinaudeau, I. (2012) ‘Action publique et partis politiques:
L’analyse de l’agenda législatif français entre 1981 et 2009’, Gouvernement et
Pinson, G. (2009) Gouverner la ville par projet: Urbanisme et gouvernance des villes
européennes, Paris: Presses de Sciences Po.
Pinson, G., Sala Pala, V. (2007) ‘Peut-on vraiment se passer de l’entretien en
sociologie de l’action publique?’, Revue française de science politique 57(5), 555–97.
Revillard, A. (2009) ‘L’expertise critique, force d’une institution faible? Le comité
du travail féminin et la genèse d’une politique d’égalité professionnelle en France
(1965–1983)’, Revue française de science politique 50(2): 279–300.
Sabatier, P.A., Jenkins-Smith, H.C. (1993) Policy Change and Learning: An Advocacy
Coalition Approach. Theoretical Lenses on Public Policy, Boulder, CO: Westview
Press.
Setbon, M. (1993) Pouvoirs contre SIDA de la transfusion sanguine au dépistage,
décisions et pratiques en France, Grande-Bretagne et Suède, Paris: Seuil.
Smith, A. (1996) Europe politique au miroir du local: Les fonds structurels et les zones
rurales en France et en Espagne, Paris: L’Harmattan.
Spenlehauer, V. (1998) L’évaluation des politiques publiques, avatar de la planification,
Grenoble: Humanities and Social Sciences, Université Pierre Mendès-France
– Grenoble II.
Thoenig, J.-C. (1976) L’ère des technocrates, Paris: L’Harmattan.
de France.
Warin, J. (1991) Les usagers dans l’évaluation des politiques publiques: Étude des relations
de service, Paris: L’Harmattan.
Woll, C. (2014) ‘Bank rescue schemes in Continental Europe: The power of
collective inaction’, Government and Opposition 49(3), 426–51.
Woll, C., Jacquot, S. (2010) ‘Using Europe: Strategic action in multi-level politics’,
Comparative European Politics 8(1), 110–26.
Zittoun, P. (2001a) La politique du logement, 1981–1995: Transformations d’une
politique publique controversée, Paris: L’Harmattan.
Zittoun, P. (2001b) ‘Partis politiques et politiques du logement, échange de
ressource entre dons et dettes politiques’, Revue Française de Science Politique 51
(October).
Zittoun, P. (2009) ‘Understanding policy change as a discursive problem’, Journal
of Comparative Policy Analysis 11(1), 65–82.
Zittoun, P. (2013) La fabrique politique des politiques publiques, Paris: Presses de
Science Po.
97
Zittoun, P. (2014) The Political Process of Policymaking: A Pragmatic Approach to

Public Policy, Basingstoke: Palgrave Macmillan.
Zittoun, P., Demongeot, B. (2010) ‘Debates in French policy studies: From
cognitive to discursive approaches’, Critical Policy Studies 3(3–4), 391–406.
98
Part Two
Policy analysis
by governments
99
SIX
Civil servants and policy analysis

in central government
Emilie Biland and Natacha Gally
Contrary to the US (since Lasswell, 1951) and Canada (Dobuzinskisk et al,

2007), French policy analysis has not emerged as a coherent body of knowledge,
methods and tools designed for improving administrative processes. This non-
institutionalisation of policy analysis can seem paradoxical in a country where
specific schools (grandes écoles de service public) are dedicated to civil service training.
Indeed, the existence of powerful Grands Corps (the most prestigious status
groups within the civil service), which have long possessed a monopoly over
the production of knowledge on and for policies, and the historical strength of
Grandes Écoles, as alternatives to universities, have most certainly contributed to
keep academics at a distance from the bureaucratic field. Therefore, administrative
knowledge of policies has been little formalised and theorised: rather than a
discipline, it consists more of practical skills and know-how that are transmitted
between civil servants, either in schools or on the job, and mobilised by the latter
in their everyday work inside departments. Thus, considering the relationship
between civil servants and policy analysis implies going beyond the nominalist
pitfall that would restrict policy analysis to knowledge and practice that is labelled
as such. Rather, this chapter will define policy analysis broadly as the legitimate
knowledge on and for policies produced and mobilised by civil servants.
Similarly, a comprehensive definition of the civil service has been chosen.
The definition of the civil service (haute fonction publique) is indeed complicated,
since higher public administration is a very segmented space, across two main
dimensions: the professional one symbolised by the existence of a great variety
of corps (each of them having a specific professional jurisdiction), and the
organisational one embodied by vertical-hierarchical (for example, junior or
senior positions), as well as horizontal-sectoral divisions (for example, between
ministers or directorates). Therefore, it would be reductionist to limit our study
to the famous Grands Corps, especially as change in civil servants’ relationship
to governing knowledge and tools has often been initiated from the civil
service margins, either by heterodox members of the Grands Corps, or even by
hierarchically or territorially less prestigious groups (middle-rank civil servants,
local officials or non-permanent staff).
Thus, this chapter will discuss the relationship between civil servants and
policy analysis in a diachronic perspective, showing which types of knowledge
have participated throughout time to what Bourdieu has called ‘pensée d’Etat’
(1993), that is, ways of thinking and classification principles that civil servants
101
produce and use to govern. It will underline the heterogeneity of this legitimated
knowledge and its evolutions since 1945, within administrative schools and in
civil servants’ daily practices.
Civil servants and policy analysis at school

Beyond the ‘generalist model’: knowledge for policies in Grandes Écoles
(1940s–1960s)
This discussion of the relationships between civil servants and policy analysis
cannot ignore higher education institutions historically designed to recruit and
educate members of the administrative elite. Indeed, one of the peculiarities of
France compared to its European neighbours lies in the existence of so-called
Grandes Écoles constituting distinct and distinguished tracks in the higher education
field and towards senior positions within the state apparatus (Bourdieu, 1989).
However, the existence of such institutions as the internationally known National
School of Administration (École nationale d’administration [ENA]) does not entail
the existence of a consistent corpus of knowledge on and for policies that would
be inculcated in higher civil servants in a uniform or specialised manner.
The field of French Grandes Écoles is indeed structured by a generalist educational
norm, historically embodied by the École Polytechnique (the state engineering
school funded in the post-revolutionary France of 1794) and progressively
deployed in other educational institutions as it defines the hierarchy of prestige
within the field (Lazuech, 1999). The administrative elite derives its ‘nobility’,
as exposed by Bourdieu (1989), from its ‘generalist culture’, whose socially
reproductive nature explains the over-representation of a white upper-class male
elite at the head of public administration and within private offices of ministers
(cabinets ministériels). Understood as the ability to mobilise general knowledge
on a wide range of topics, as a skill to synthesise complex situations, and as an
aptitude to take various positions within – and even outside – government, the
unequal distribution of this attitude toward policy knowledge is partly due to
the stratification of the civil service. Whereas specialisation is conceived as an
obstacle to successful career strategies in higher administration (Suleiman, 1979),
the study of the ENA training in the 1990s has shown the predominance of
formal know-hows and reproductive socialisation mechanisms in the shaping of
administrative elites, focused on the transmission of ‘appropriate’ ways of thinking
and ‘conforming’ behaviours rather than substantial knowledge (Eymeri, 2001,79).
Overall, specialist or technical knowledge remains associated with ‘dirty work’
(Hughes, 1962) and has historically been reported on subordinate staff such as
attachés (middle-rank officials) (Quéré, 2015; Quéré 2017) in order to preserve
the higher civil service monopoly on prestigious activities of political advice and
policy conception.
This ‘generalist model’ does not, however, exhaust the question of civil servants’
socialisation to policy knowledge. First, it does not apply to the entire civil
102
Civil servants and policy analysis in central government
service: notably, it does not account for specific knowledge and know-how at the
core of the historical institutionalisation process of engineering Grandes Écoles,
such as the École des Ponts et Chaussées (School for Bridges and Roads) created
in 1747 in order to train competent officers for bridges and roads. Unlike their
administrative counterparts, state engineering Grands Corps have historically been
trained in specialised subjects according to the career they chose (Mines; Bridges
and Roads; Civil Aviation; Statisticians, and so on) in one of Polytechnique’s post-
graduate schools for practical training (Écoles d’application). Second, the ‘generalist’
category is indeed ambiguous and hides the successive transformations of the
ENA curriculum (Biland and Kolopp, 2013; Biland and Vanneuville, 2012).
Therefore, studying the relationships between civil servants and policy analysis
requires exploring the variety of meanings the ‘generalist model’ has taken over
time in specific administrative schools.
First, the creation of a generalist school to educate civil servants in 1945
appeared as an ‘historical compromise’ (Chagnollaud, 1991, 149) between moral,
democratic and professional concerns. This reform of the administrative elite’s
education had a symbolic character, designed to restore the civil service’s image
after the collaborationist Vichy episode. Indeed, the former École libre des sciences
politiques (a private elite higher education institution founded in Paris in 1870)
was nationalised and renamed the Institute for Political Studies (Institut d’Etudes
Politiques). Along with the creation of similar institutes in several cities and of
the ENA, this change intended to give the future elite a better sense of ‘public
service’ and ‘general interest’ while opening the higher civil service to middle-
class students. However, the ENA founding fathers’ idea was also to foster the
emergence of a unified and professional senior civil service, properly trained to
administrative matters and specialised according to four programmes, matching
four main types of careers: general administration, economic and financial
administration, social administration and diplomacy. Hence, topics taught at the
ENA were simultaneously general and specialised, moral and professional, these
two facets being constitutive of the ambiguous agreement of 1945.
Biland and Kolopp (2013) have shown this major tension between a ‘humanistic’
and a ‘scientific’ conception of higher civil servants’ education. Whereas the ENA
was originally oriented towards general and abstract knowledge and dominated by
public law (Biland and Vanneuville, 2012, 34), the humanistic curricula have been
criticised early on as too far from practical issues. From the mid-1950s onward,
this ‘humanistic’ model eroded as classes were reoriented towards economics –
especially quantitative/mathematical economics – and became less encyclopedic
and more case-oriented (Biland and Kolopp, 2013, 241–3). This turn was also
illustrated by the nature of entrance examinations, which were progressively
oriented towards policy problems and technical case studies from 1958 onwards
(Mangenot, 1999). Reflecting this further detachment from the university
reference, the number of academics in the administrative board of the school
decreased steadily, while a quasi-permanent pedagogic team – mainly composed
of civil servants – emerged at the margin of the board (Biland and Kolopp,
103
2013, 227–30). In the meantime, leading officials of the General Commission

for Planning (Commissariat général du Plan [CGP]) and of the National Institute
for Statistics and Economic Studies (Institut national de la statistique et des études
économiques [INSEE]), teaching economics in a problem solving and ‘modernising’
perspective, turned out to be central figures of the school (Kolopp, 2013).
Concomitantly with this shift towards economics and specialisation, the
organisation of teaching at the ENA became paradoxically less sector-oriented,
since the 1958 reform abolished the four programmes mentioned earlier.
Resulting from a process of gradual change related to the concern of civil servants’
professional organisations – the Grands Corps especially – for controlling their
own recruitment and maintaining the implicit hierarchy associated with the
generalist norm (Gally, 2012, 259–68), this reform entailed that knowledge of
and for policies was conceived as cross-sectoral, consistently with the centralist
planning and programming ideology of the CGP (Pasquier, 2003). A similar
shift was observable within engineering state schools, at least within the School
for Bridges and Roads (Gervais, 2007a), where teaching increasingly turned
towards ‘social sciences’ or ‘humanities’ from the 1960s onwards. Indeed, the
education of state engineers, traditionally focused on scientific and technical
matters linked to structural engineering (for example, ‘resistance of materials’,
‘reinforced concrete’, ‘steel construction’), has included increasingly more law
or economics, in the form of elective courses followed by second-year students
(Gervais, 2007a, 391–2). The increasing importance of courses considered as
‘non-technical’ has hybridised the historical curriculum of the school based
on knowledge and skills specific to a policy sector (that is, public works). This
allowed further competition with ENA alumni for the access to senior positions
within and outside higher administration. In the 1960s the influence of the
‘generalist model’ was also illustrated by the creation of the National Centre for
Judicial Studies (Centre national d’études judiciaires [now known as École nationale
de la magistrature]) in order to educate future judges as ‘honest men’ and generate
interest for non-judicial techniques such as forensic medicine or accounting
(Boigeol, 2013, 19–22).
Overall, the ‘generalist norm’ appears as a cornerstone in the field of elite schools
as well as the cognitive basis for male domination over public administration:
till 1976, all members of the ENA board were men; by the mid-1960s, very
few women had reached the most prestigious and lucrative positions in the civil
service (2 per cent of women among top officials in the Finance department in
1964, for instance) (Darbel and Schnapper, 1969, 126). Nevertheless, this male
dominance should not overshadow the heterogeneity of civil servants’ relationships
to knowledge on and for policy. Policy analysis, as it is inculcated in senior officials
embraces a wide diversity of subjects and teaching methods, which have varied
among different schools and throughout time. This diversity is indeed directly
related to the intern segmentation of the higher civil service, consisting of a
plurality of professional groups in competition for positions within and outside
the state (Gally, 2012). In this perspective, knowledge is undoubtedly a resource
104
likely to be used to claim new professional jurisdictions and (re)negotiate the

professional boundaries of the higher civil service.
Managerial shifts in civil servants’ training (1970s–2010s)
Since the 1970s, changes in these training patterns have been neither radical nor
sudden. Prestigious schools of the civil service as well as powerful status groups
have subsisted. Policy analysis has not been institutionalised as an autonomous
field of knowledge. Nevertheless, along with the diffusion of managerial principles
within government (Bezes, 2009), curricula in several Grandes Écoles have shifted
towards a managerial perspective on policy making (Biland and Kolopp, 2013).
This trend has had diverse paths depending on the school. As the leader of the
French civil service vanguard, the board of the ENA has long been reluctant to
such a change. During the 1970s, professors of law and members of the Council
of State (Conseil d’État), which stands as the highest administrative court, have
postponed in this school the implementation of management classes, which they
devalued as a ‘set of tools’ (Bezes, 2012).
As a result, managerial training has first been implemented in less prestigious
institutions. Continuing education programmes for administrators (Bezes, 2012),
engineers (Chanut, 2001) and public health officials (Buton and Pierru, 2012)
have offered a setting at the margins of schools to experiment the teaching of
new knowledge. Within the local civil service, administrators used management
science as a way to upgrade their status (traditionally far below that of the ENA
alumni’s) while distinguishing themselves from lower local civil servants. Designed
in 1984 as a continuing education programme, the training centre for senior
managers (Centre Supérieur de Formation des Cadres) was described as halfway
between state schools and business schools (Roubieu, 1999). As a matter of
fact, it involved consultants as well as scholars specialised in management, who
were familiar to the US business programmes – some of them having failed to
implement such a programme at the ENA. This training centre has legitimised
a new professional figure: favouring abilities regarding policy outcomes (rather
than internal procedures) to achieve efficiency (Thoenig, 1994), this has helped
local civil servants to get influence over elected officials (Roubieu, 1994).
In state schools, management science has risen in comparison to the ENA
‘generalist’ model. In the Regional Institutes for Administration (Instituts Régionaux
d’Administration), which train middle-rank officials, management has been taught
since these schools were established, back in 1970. It was first thought of as a
technical tool (such as ‘budgeting techniques’ and ‘managerial accounting’), thus
fitting in with these officials’ subordinate position. It has been gradually used,
however, to legitimate their ability to design public policy and to be involved
in decision-making, in other words to be granted autonomy from the ENA
alumni (Quéré, 2014). At the School for Bridges and Roads, the teaching of
management began in the 1970s with a new course designed for both engineering
and business students (the latter being registered at HEC, the most well-known
105
French business school). Dual career opportunities for alumni (either in private
companies or in public administration) encouraged this change. Beginning in the
1980s, the transfer of powers and skills in Public Works towards local authorities
then encouraged these State engineers to endorse a less technical but more
managerial professional identity. Management teaching helped adapt students
to new career opportunities – within and outside the state – and finally to get
material and symbolic benefits (Gervais, 2007b).
As compared to these schools, has the ENA curriculum remained on the
sidelines of this managerial shift? In fact, its initial reluctance has faded away
since the 1980s. The first class of management was circumspectly implemented
in 1982, but, since the 2006 reform, the ‘public management’ training module
has been the core of the curriculum and has included a 15-week internship in
private companies. In 2015, according to the ENA website, its goal is ‘to turn
students into managers who are able to manage teams and lead projects’. Such
an evolution has not made a clean slate of the past though: to put it briefly, legal
and economic knowledge are still important in the ENA curriculum. Rather, it
has been possible because most of the powerful status groups in this school have
believed that they could benefit from it. The conversion of jurists to management
teaching is crucial, since they used to strongly oppose it. Threatened by domestic
rivals (Grands Corps specialised in economic and financial matters) as well as
by international institutions (especially the European courts), members of the
Council of State have gradually reshaped legal teaching at the ENA in order to
counter criticisms of being ‘irrelevant’ to ‘modern’ government. Since the 2000s,
legal classes have focused on the teaching of ‘legistics’, which is namely the art of
drafting proper legal texts, as well as on the jurists’ plea for the political usefulness
and effectiveness of law, as a toolbox complying with economic rationality
(Biland and Vanneuville, 2012). In short, such an ‘instrumentalisation’ of law
(Morand, 1999) is the result of jurists’ efforts to keep influence and power within
government, in the age of new public management (NPM).
At last, management training is quite heterogeneous from one school to the
other and it has been implemented through rather diverse timings. This should
prevent scholars from considering it as an overwhelming wave in civil servants’
education. Indeed, the rise of managerial education results from struggles between
status groups and between public institutions. It has come from actors and schools
at the margins of historical state prestige and has then been undertaken by senior
players whose power was threatened by diverse trends (from decentralisation to
critics of legal expertise). Further, such a trend cannot be understood from an
exclusively state standpoint: these decades are characterised by growing interactions
between public and private actors, as well as between domestic and international
ones. Managerial training has involved consultants and foreign inspirers. It has
favoured business-trained students’ careers in the public service, which is linked
to the rise of civil servants taking jobs in private companies. This practice of
pantouflage has developed during this period (Rouban, 2010); it has encouraged
the teaching of pieces of knowledge transferable from private companies to
106
public administration and vice versa (Biland and Kolopp, 2013). Overall, the
consequences of this managerial shift on the redefinition of sources of legitimate
expertise remain questionable. Are managers just old-fashioned generalists in new
clothes? Or does the managerial toolbox represent an opportunity for alternative
or formerly dominated actors in the civil service? Recent research has pointed
out the reconfiguration of power relationships within professional bureaucracies
(Barrier and Musselin, 2015) and administrative corps (Boussard and Loriol, 2008).
However, the effects of NPM over the gendered division of administrative labour
remain ambivalent (Bereni et al, 2011), as illustrated by the persistant ‘glass ceiling’
for female employees within the French higher civil service (women counted for
about a quarter of top officials in 2014) (Jacquemart et al, 2016).
Civil servants and policy analysis at work

The historical diversification of policy knowledge production
Whereas civil service schools are undoubtedly critical institutions for the
production of policy analysis, expertise is also developed and mobilised by
various – and often competing – administrative organisations whose importance
has evolved over time. While the production of policy expertise within French
administration was historically related to the power of the Grands Corps, their
monopoly over knowledge on and for policies has been seriously challenged
after the Second World War by new centralist organisations involved in planning
and programming activities. Under the fifth Republic, ministerial cabinets have
emerged as a crucial source of policy analysis, these positions being practically
monopolised by higher civil servants. At the departmental level, the development
of statistical services represents another resource for the production of sectoral
expertise.
Policy analysis has been historically produced through a division of labour
between technical and administrative Grands Corps. On the one hand, the
former have based their legitimacy upon sectoral policy expertise mobilised in
the conception of public policies such as road extension or the regulation of
mines since the end of the eighteenth century. On the other hand, the Council
of State, the Court of Auditors (Cour des comptes) and the General Finance
Inspectorate (Inspection générale des finances) were created in the Napoleonic
era and the Restoration period as centralist organisations located outside
departmental hierarchy, examining administrative activities downstream of the
policy process (Bodiguel and Quermonne, 1983; Kessler, 1986). However,
these initial boundaries have progressively been blurred: administrative Grands
Corps have developed their function of policy advice ahead of political decisions
and/or in specific policy sectors, while engineering corps have claimed a cross-
sectoral competence. The Council of State has been increasingly considered as
a source of expertise on public policies: its thematic annual reports focused on
specific policy issues (‘agencification’ policies, citizens’ participation to public
107
decision, water regulation, and so on) have been increasingly influential, along
with the strengthening of its research department (section des études) (Biland and
Vanneuville, 2014). In the social sector, the Court of Auditors and the General
Inspectorate for Social Affairs (Inspection générale des affaires sociales [IGAS]) have
also developed prospective sectoral expertise and serve as socialisation places for an
emergent policy elite since the 1980s (Genieys and Hassenteufel, 2001; Hervier,
2008; Genieys, 2010; Genieys and Hassenteufel, 2012). In the meantime, state
engineers have invested other type of knowledge as those directly related to
their policy sector. For instance, some engineers of Bridges and Roads claimed
to be competent in economics within the former Department of Public Works
(Mazoyer, 2012; 2013). From 1995 to 2015, several state engineering corps have
merged, so that their number decreased from eleven to five, thus questioning the
organic link between corps, sectoral departments and technical sector-oriented
knowledge (Gervais, 2011).
In addition to these evolutions in Grands Corps, alternative sources of policy
expertise have emerged in relationship to the transformations of state intervention.
In the post-war period and along with the development of planning and
programming activities, centralist organisations, called administrative task forces
(administrations de mission) (Pisani, 1956), have been created outside bureaucratic
hierarchy and with non-permanent staff (that is, civil servants on secondment
or non-civil servants on fixed term contracts). The most important of these was
certainly the General Commission for Planning (CGP), which was designed in
1946 to direct and steer the French economy in the reconstruction period. The
first generation of CGP members had a heterodox profile compared to sectoral
departments’ civil servants and members of the Grands Corps. They had been
socialised outside public administration – particularly in the banking sector – and
in the Resistance besides the General De Gaulle (Pasquier, 2003, 108). They
first served as mediators between French and American administrations for the
management of the Marshall Plan, and imported national accounting principles
(Spenlehauer, 1999). Then, they developed a role of ‘modernisers’, spreading out
macro-economic theories and defending a strong-willed conception of economic
policy and state intervention (Gaïti, 2002). De Gaulle’s return to power in 1958
and the context of economic growth favoured the extension of the CGP’s remit
in the 1960s, from a few policy sectors narrowly linked to post-war reconstruction
such as charcoal, electricity, agricultural machinery, or transport infrastructures,
to general matters related to economic and social development. For almost 50
years, the CGP, which reported directly to the Prime Minister, played a crucial
role in the production of expertise on public policies and as a consultative forum
whose five-year plans were approved by Parliament and had formal legal force.
Starting from the middle of the 1980s though, it became increasingly criticised,
along with the overall questioning of the state. Its planning function was
abandoned in 1993 and gradually refocused on prospective studies. Concomitantly,
several other committees reporting directly to the Prime Minister were created
to tackle specific sectoral subjects such as economics (Conseil d’analyse économique
108
since 1997) and pensions (Conseil d’Orientation des Retraites since 2000). In 2005,
the CGP was renamed Centre for Strategic Analysis (Centre d’analyse stratégique)
and reorganised along the model of the UK Prime Minister’s Strategy Unit, as
the coordinator of the centralist committees’ network (Tirera, 2005, 23). It has
become the General Commission for Strategy and Prospective (Commissariat
general à la stratégie et à la prospective) in 2013, and is currently in charge of advising
the government on various policy issues in addition to its already mentioned
coordination role.
In the meantime, departments have played an increasing role in the production
of knowledge on and for policies. Compared to other countries, a crucial
feature of the Fifth Republic politico-administrative system is the existence of
powerful ministerial cabinets, mainly staffed with civil servants. Working at the
crossroads of policy and politics, they provide politicians with policy knowledge
coming up from departments, and conversely adapt policy instruments to match
issues of timing and political feasibility (Eymeri, 2003). In addition, the role
of departmental inspection corps must not be overlooked. Besides three cross-
departmental inspections (the already mentioned IGF and IGAS, plus the General
Inspection of Administration [Inspection générale de l’administration]), these are
important knowledge producers for their respective departments. Historically
focused on administrative control, some of them have recently been reformed,
such as the Economic and Financial General Control (Contrôle général économique
et financier), and their role extended to audit, evaluation, research and policy
advice (Coppolani, 2007). At last, departments have also encouraged the creation
of directorates dedicated to the production of knowledge on and for policies,
especially statistical data and performance indicators. The most famous of these is
certainly the already mentioned INSEE, created in 1946 as a general directorate of
the Finance Department but whose organisational autonomy was acknowledged
in 2008. While it does not directly take part in public policies, the INSEE remains
nowadays a key producer of statistical data on French society, and therefore
contributes to shape the categories through which policies are formulated and
implemented (Desrosières, 1993). Ministerial Statistical Services (MSS) have also
played an increasing role in the publication of sectoral knowledge likely to be
mobilised in policy formulation. The Research and Statistics Unit (Direction de
l’animation de la recherche, des études et des statistiques) (DARES) at the Department
of Work and Employment and the Research, Evaluation and Statistics Unit
(Direction de la recherche, des études, de l’évaluation et des statistiques) (DREES) at the
Department of Public Health are the most well-known MSS. Mainly staffed with
middle-rank civil servants, these statistical services show that the production of
policy knowledge in central government is not the monopoly of administrative
elites (Penissat, 2012). The creation in 1965 of the Committee of Women’s Work
(Comité du travail féminin) (inspired by the US Women’s bureau) at the Department
of Work and Employment further illustrates the non-monopoly of higher civil
servants over the production of policy expertise: the institutionalisation of State
109
feminism in France has led to the administrative acclimatisation of knowledge

on and for women and gender equality (Revillard, 2016, 50–60).
The Division for Economic and Financial Studies (Service des études économiques
et financières) of the Finance Department is another famous example of
departmental source of policy knowledge. Created in 1947 within the Treasury
directorate of the Finance Department, it was institutionalised as an autonomous
Prospective Directorate (Direction de la prévision) in 1965. It played a crucial
role in the emergence of a reflexive body of knowledge over administrative
activities, especially costs and productivity indicators, in the framework of the
planning-programming-budgeting system (the Rationalisation des choix budgétaires
programme) (Bezes, 2009, 68–71). At the same period, other policy-oriented units
were created within various departments, for example within the Department
of Public Works, where the role of the Division of Economic and International
Affairs (Service des affaires économiques et internationales [SAEI]) has been recently
documented (Mazoyer, 2012; 2013). More recently, the rise in power of the Social
Security Directorate (Direction de la sécurité sociale) is another example of how
sectoral expertise can serve as a resource to negotiate with the powerful Budget
Directorate (Direction du Budget) of the Finance ministry (Hassenteufel, 2012).
To put it briefly, organisations involved in the production of policy analysis
have considerably increased in number and changed in nature over time. Moving
from organisations to individuals, how is policy analysis actually mobilised in
professional practice?
Practicing policy analysis
First of all, the institutionalisation of a profession devoted to ‘policy analysis’ within

the French public administration has never really occurred. Created in 2010, the
on-line job market for central government (Bourse interministérielle de l’emploi public)
indexes 28 professional areas, two being explicitly linked to public policy. In May
2017, 443 job ads (out of 7,713) were indexed in the area called ‘Development
and management of public policies’ and targeted senior civil servants; 204 ads were
associated to the ‘evaluation of public policies’; most of them asked for middle-
rank civil servants. These ads indicate the growing acknowledgment of ‘policy
analysis’ vocabulary within public administration. However, this trend has not
reached the same degree as in several other western countries where ‘policy analyst’
is a dignified governmental job. In France, such labels are not autonomous; they
gather various jobs, which have little in common except highlighting strategic
tasks rather than operational ones and targeting qualified professionals. To analyse
what kind of knowledge and skills civil servants use in their daily practice, one
shall not restrict the scope to these labels, but rather include broader professional
categories, from Grands Corps to middle-rank civil servants.
For the last 15 years, several studies have used a sociological approach to
investigate civil servants at work. They show that the latter’s writing skills are key
resources to act as policy advisers: drafting documents – from notes and reports
110
to regulations and bills – takes a major part in civil servants’ work, since law has
long been granted the status of ‘official thought’ within the French government
(Caillosse, 1998, 81). When they take a new position, the socialisation to their
new working environment includes learning how to write properly (Laurens,
2013): this ‘learning on the job’ process is indeed crucial to incorporate the ‘house
culture’ (Eymeri, 1999), which differentiates one department from the others
regarding the kind of policy knowledge that is mobilised. Drafting is indeed the
main vehicle to convince elected officials, that is, to frame political debates and
solutions (Laurens, 2013) and to define which matters are political – requiring
the arbitration of the minister – and which matters are technical – being under
the administrative staff’s jurisdiction (Eymeri, 2003). Writing texts is also a means
to get influence outside government, since those documents latter qualify public
issues and may suggest policy proposals. Parts of confidential notes may be turned
into public discourses to the media (Laurens, 2013). Moreover, since the 1980s,
the MSS issue publicly available short papers, which have been used by statisticians
to inform public opinion on government policies (Penissat, 2012).
Concomitant with their growing influence within public administration, this
involvement of MSS in publishing activities is part of a major trend to ‘turn
words into numbers’ (Desrosières, 2008) in order to quantify social and economic
issues along with public policies. Certainly, quantification is not the one and
only dimension of NPM; besides, this tendency appeared prior to the NPM
period. As mentioned earlier, the planning-programming-budgeting system
stimulated quantitative economic expertise and legitimated the new figure of
‘economist engineer’ during the 1960s and 1970s (Dulong, 1996; Mazoyer,
2012). Nevertheless, the diffusion of quantification in every sector of public
policy, far beyond economic and technical departments, is attributable to the
growing influence of managerial ideas within government. Formerly a self-
regulated sector, valuing slowness as a quality guarantee, the Justice department
has faced major changes since the 1990s, combining a lack of resources and the
rise of quantitative objectives, such as reducing judicial costs and delays in Court
(Vigour, 2006; Vauchez, 2008). In the education sector, pupil flows have been
quantified since the 1990s, in order to arrange academic orientation and to
allocate resources between schools (Normand, 2011). Since the 2000s, this trend
has gained ground in the higher education sector, which is targeted by European
policies (Bologna declaration in 1999, Lisbon European Submit in 2000). It has
led to the creation of new agencies in 2006–07: The French National Research
Agency (Agence Nationale de la Recherche) for funding; and the Research and
Higher Education Evaluation Council (Haut Conseil de l’évaluation de la recherche
et de l’enseignement supérieur) for evaluation. The latter undertakes benchmarking
activities, regarding universities and scholars (Bruno and Didier, 2013). In this
period of scarce resources, these agencies encourage programmes from which
private companies could benefit. Finally, the healthcare sector is probably the one
most affected by this managerial shift. Competition between hospitals is organised
by several agencies: National Performance Support Agency (Agence nationale
111
d’appui à la performance) and National Authority for Health (Haute Autorité de

Santé) nationwide; Regional Health agencies (Agences Régionales de Santé) in each
region. This competition relies on financial mechanisms based on performance
(T2A standing for Price for Activity) rather than on inputs (Pierru, 2007).
To what extent do civil servants support the use of such quantitative indicators?
First promoted by State economists in order to enforce budgetary restraint
(Jobert and Théret, 1994), these have then been appropriated by senior officials
involved in other sectors. During the 1990s, especially, the ‘welfare elite’ (that
is, the members of Grands Corps who got involved in social policies) pleaded for
the budgetary equilibrium of the welfare system (Sécurité Sociale) in order to limit
the interference of economic administrators as well as the traditional influence
of doctors in this sector (Genieys and Hassenteufel, 2012). In local welfare
offices, attitudes towards managerialism are more ambiguous. Among social
administrators of conseils départementaux (local public administration in charge of
social policies), ‘managerial language’ (for instance, the use of the term ‘social
performance’) is widespread. However, there is no common knowledge about
management principles. What administrators have in mind when they use these
terms varies a lot depending on their professional and educational background.
Social administrators trained at the ENA or at the INET (the training institute
for local higher officials), who have worked in different sectors, are more likely
to value metrics on social policies than social workers who have climbed the
bureaucratic ladder within the welfare sector (Alcaras et al, 2014).
Generally, policy knowledge is at stake in many conflicts between State elites.
In 2007, for example, statisticians from the DARES went on strike against the
publication of the ‘official unemployment figures’ that they considered misleading
and badly designed. They defended the autonomy of statisticians in producing
quality against political interference, while criticising the ineffectiveness of the
right-wing government’s policy on unemployment (Penissat, 2012). The self-
regulation of administrators in designing indicators has also been contested during
the 2000s regarding environmental indicators. First sustainable development
indicators (SDIs) were designed within a small independent agency, the French
Institute for the Environment (Institut Français de l’Environnement). Though inspired
by the United Nations’ SDIs, those French ones were more policy-oriented.
In 2003, the right-wing government devalued them as too ‘political’ (meaning
‘environmentalist’) and decided to internalise the production of indicators within
the Department of Environment in order to control them closely. This result is
a shift ‘to a minimalist view of the role of SDIs in policy making’ (Le Bourhis,
2015, 11). Gender equality indicators, which have been developed as criteria of
evaluation in many policy sectors since the Organic Law on Finance Laws (Loi
organique relative aux lois de finances [LOLF]) passed in 2001 (Mazur, 2007), are a
third illustration of the political and conflictual use of performance indicators.
Gender equality as a cross-sectoral policy goal (known as ‘gender mainstreaming’)
undermines the capacity of feminist activists to shape and control these instruments
in public administration (Jacquot, 2015; Revillard, 2016).
112
These examples should prevent observers from concluding on the irrefutable

impact of indicators on policy analysis. The institutionalisation of public policy
evaluation within public administration is worth being summed up in order to
assess this impact. Evaluation practices were dispersed, slightly visible and poorly
used in policy-making until the end of the 1990s (Perret, 1999). During the
2000s, the LOLF (which reformed the budget-making process) and the General
Public Policy Review (Révision générale des politiques publiques) (that led to 170
modernisation audits), were the two main programmes for State modernisation.
Both aimed at developing the culture of evaluation, but they failed to achieve it.
First, policy evaluation is less an administrative practice than a market for private
consulting firms (see Gervais and Pierru in this volume). Second, the financial
goal of these programmes (finding quick and substantial budgetary savings) took
precedence over the rigorous evaluation of policy achievement. Despite the
plea for participative evaluation at the end of 1980s (Viveret, 1989), evaluation
indicators still focus on metrics. Certainly, policy evaluation as practised by civil
servants is often restricted to a quantitative and financial approach. This narrow
definition may be an impediment to the development of comprehensive evaluation
practices, as they are valued in the international community of evaluators
(Lacouette Fougère and Lascoumes, 2013).
Conclusion
This chapter has pointed out the persistent prevailing role of top civil servants
in the production and mobilisation of policy analysis in the French context. Top
officials have historically built their legitimacy on the monopolistic detention of
policy knowledge, transmitted within administrative Grandes Écoles, and mobilised
at the central or ministerial level. This monopolistic situation certainly contributes
to an explanation of one of the specificities of the French case, which is the
gap between policy analysis as it is theorised by academics within universities
(see Hassenteufel and Le Galès in this volume) and knowledge for policies as it
is developed within government. Further, the diagnosis of a ‘generalist ethos’
is helpful as compared to other national contexts. However, it should not hide
the diversity of knowledge mobilised by the civil service throughout time and
sectors. What policy analysis means and how it is practised within the state
depends on the internal competition for the definition of legitimate knowledge
between various segments of the civil service. The diffusion of neo-managerial
tools since the 1980s should be analysed in relation to these administrative
struggles. Though not specific to the French case, two significant evolutions are
the increasing influence of ‘numbers’ compared to more traditional literary or
legal skills and the shift of policy expertise downstream of the policy process, as
top civil servants’ work has been increasingly oriented toward policy evaluation
and performance measures. Therefore, the study of administrative elites’ role
in producing and using policy knowledge is now more than ever necessary to
understand public policies (Page, 2012). Their ability to master these new types
113
of policy knowledge certainly conditions the persistence of their power over

rival actors, such as consultants, private-sector experts and members of interest
groups, who strengthen as alternative sources of knowledge on and for policies.
References
Alcaras, J.-R., Marchand, C., Marrel, G., Nonjon, M. (2014) ‘Quand les directeurs
des services départementaux se saisissent du management: des représentations
et des pratiques singulières et équivoques’, Politiques et management public 31(3),
337–56.
Barrier, J., Musselin, C. (2015) ‘La réforme comme opportunité professionnelle?
Autonomie des établissements et montée en puissance des cadres administratifs
des universités’, Gouvernement et action publique 4(4), 127–51.
Bereni, L., Marry, C., Pochic, S., Revillard, A. (2011) ‘Le plafond de verre dans les
ministères: Regards croisés de la sociologie du travail et de la science politique’,
Bezes, P. (2009) Réinventer l’État: Les réformes de l’administration française (1962–
2008), Paris: Presses Universitaires de France.
Bezes, P. (2012) ‘État, experts et savoirs néo-managériaux: Les producteurs et
diffuseurs du New Public Management en France depuis les années 1970’, Actes
Biland, E., Kolopp, S. (2013) ‘La fabrique de la pensée d’État: Luttes d’institutions
et arrangements cognitifs à l’ENA (1945–1982)’, Gouvernement et action publique
2(2), 221–48.
Biland, E., Vanneuville, R. (2012), ‘Government lawyers and the training of senior
civil servants: Maintaining law at the heart of the French state’, International
Journal of the Legal Profession 19(1), 29–54.
Biland, E., Vanneuville, R. (2014) ‘Les mutations de la surveillance juridique des
administrations: Le conseil d’État français et la promotion de la légistique’, in
J. Crête (ed) Les surveillants de l’État démocratique, pp.109–33, Québec: Presses
de l’Université Laval.
Bodiguel, J.-L., Quermonne, J.-L. (1983) La haute fonction publique sous la Ve
République, Paris: Presses Universitaires de France.
Boigeol, A. (2013) ‘Quel droit pour quel magistrat? Évolution de la place du droit
dans la formation des magistrats français, 1958–2005’, Droit et société 83, 17–31.
Bourdieu, P. (1989) La noblesse d’État: Grandes écoles et esprit de corps, Paris: Minuit.
Bourdieu, P. (1993) ‘Esprits d’État: Genèse et structure du champ bureaucratique’,
Actes de la recherche en sciences sociales 96–7, 49–62.
Boussard, V., Loriol, M (2008) ‘Les cadres du ministère des affaires étrangères et
européennes face à la LOLF’, Revue française d’administration publique 128(4),
717–28.
Bruno, I., Didier, E. (2013) Benchmarking: L’État sous pression statistique, Paris:
La Découverte.
114
Buton, F., Pierru, F. (2012) ‘Instituer la police des risques sanitaires: Mise en
circulation de l’épidémiologie appliquée et agencification de l’État sanitaire’,
Gouvernement et action publique 4, 67–90.
Caillosse, J. (1998) Introduire au droit, Paris: Montchrestien.
Chagnollaud, D. (1991) Le premier des ordres: Les hauts fonctionnaires XVIIIe–XXe
siècle, Paris: Fayard.
Chanut, V. (2001) L’État employeur et l’État didactique: Politique de gestion et formation
au management public des cadres supérieurs du ministère de l’équipement, PhD thesis,
Paris: Université Paris 1.
Coppolani, C. (2007) ‘Le contrôle général économique et financier’, Revue
française d’administration publique 124(4), 625–31.
Darbel, A., Schnapper D. (1969) Les agents du système administratif, Paris: Mouton.
Desrosières, A. (1993) La politique des grands nombres: Histoire de la raison statistique,
Paris: La Découverte.
Desrosières, A. (2008) Gouverner par les nombres: L’argument statistique II, Paris:
Mines ParisTech.
Dobuzinskis, L., Howlett, M., Laycock, D. (eds) (2007) Policy Analysis in Canada,
Toronto: University of Toronto Press.
Dulong, D. (1996) ‘Quand l’économie devient politique: La conversion de la
compétence économique en compétence politique sous la Ve République’,
Politix 9(35), 109–30.
Eymeri, J.-M. (1999) Les gardiens de l’État: Une sociologie des énarques de ministère,
PhD thesis, Paris: Université Paris 1.
Eymeri, J.-M. (2001) La fabrique des énarques, Paris: Economica.
Eymeri, J.-M. (2003) ‘Frontière ou marches? De la contribution de la haute
administration à la production du politique’, in J. Lagroye (ed) La politisation,
pp. 47–77, Paris: Belin.
Gaïti, B. (2002) ‘Les modernisateurs dans l’administration d’après-guerre:
L’écriture d’une histoire héroïque’, Revue française d’administration publique
102(2): 295–306.
Gally, N. (2012) Le marché des hauts fonctionnaires: Comparaison des politiques de la
haute fonction publique en France et en Grande-Bretagne, PhD thesis, Paris: Institut
d’Études Politiques.
Genieys, W. (2010) The New Custodians of the State: Programmatic Elites in French
Society, New Brunswick, NJ: Transaction Publishers.
Genieys, W., Hassenteufel, P. (2001) ‘Entre les politiques publiques et la politique:
l’émergence d’une “élite du Welfare”?’, Revue française des affaires sociales 4,
41–50.
Par-delà la sociologie des élites’, Gouvernement et action publique 2(2), 89–115.
Gervais, J. (2007a) La réforme des cadres de l’action publique ou la fabrique d’un ‘nouveau’
corps des Ponts et Chaussées. Impératifs managériaux, logiques administratives et stratégies
corporatistes (fin du XXème siècle), PhD thesis, Lyon: Université Lumière Lyon 2,
Institut d’Études Politiques.
115
Gervais, J. (2007b) ‘Former des hauts fonctionnaires techniques comme des

managers de l’action publique. L’“identité managériale”, le corps des Ponts et
Chaussées et son rapport à l’Etat’, Politix 20(79), 101–23.
Gervais, J. (2011) ‘The rise of managerialism as a result of bureaucratic strategies
and power games’, in J.-M. Eymeri-Douzans, J. Pierre (eds) Administrative
Reform, Democratic Governance, and the Quality of Government, pp. 80–93, London:
Routledge.
Hassenteufel, P. (2012) ‘Les sources intellectuelles des réformes du système de
santé français: la prédominance des forums et des acteurs administratifs’, in J. de
Kervasdoué (ed) Carnet de santé de la France 2012, pp. 161–83, Paris: Economica.
Hervier, L. (2008) ‘Le rôle des organismes de contrôle en matière d’évaluation.
1949–2007: l’exemple de la Cour des comptes’, Informations sociales 150(6),
44–55.
Hughes, E.C. (1962) ‘Good people and dirty work’, Social Problems,10(1), 3–11.
Jacquemart, A, Le Mancq F., Pochic S. (2016), ‘Femmes hautes fonctionnaires en
France: L’avènement d’une égalité élitiste’, Travail, genre et sociétés 35(1), 27–45.
Jacquot, S. (2015), Transformations in EU Gender Equality: From Emergence to
Dismantling, Basingstoke: Palgrave Macmillan.
Jobert, B., Théret, B. (1994) ‘France: La consécration républicaine du néo-
libéralisme’, in B. Jobert (ed) Le tournant néo-libéral en Europe, Paris: L’Harmattan.
Kessler, M.-C. (1986) Les grands corps de l’Etat, Paris: Presses de la FNSP.
Kolopp, S. (2013) ‘De la modernisation à la raison économique: La formation en
économie à l’Ena et les déplacements des lieux communs de l’action publique
(1945–1984)’, Genèses 93(4), 53–75.
Lacouette Fougère, C., Lascoumes, P. (2013) ‘L’évaluation: Un marronnier de
l’action gouvernementale?’, Revue française d’administration publique 148(4),
859–75.
Lasswell, H. (1951) ‘The Policy Orientation’, in D. Lerner and H. Lasswell, The
policy sciences: Recent developments in scope and method, pp. 3-15, Stanford: Stanford
University Press.
Lasswell, H.D. (1968) ‘The policy sciences’, in R.K. Merton, D.L. Sills (eds)
International Encyclopedia of the Social Sciences 12, pp. 181–9, New York:
Macmillan and The Free Press.
Laurens, S. (2013) ‘Une sociologie de la note administrative: Notations sur un
instrument du pouvoir administratif ’, in J.-M. Eymeri-Douzans, G. Bouckaert
(eds) La France et ses administrations: Un état des savoirs, pp. 379–410, Brussels:
Bruylant-DeBoeck.
Lazuech, G. (1999) L’exception française: Le modèle des grandes écoles à l’épreuve de
la mondialisation, Nantes: Presses universitaires de Nantes.
Le Bourhis, J.-P. (2015) ‘The politics of green knowledge: A comparative study
of support for and resistance to sustainability and environmental indicators’,
Journal of Comparative Policy Analysis: Research and Practice, 18(4), 1–16.
116
Mangenot, M. (1999) ‘‘‘L’entrée en technocratie” : Le concours de l’ENA et les

transformations du modèle du haut fonctionnaire’, in D. Dulong, V. Dubois
(eds) La question technocratique: De l’invention d’une figure aux transformations de
l’action publique, pp. 93–107, Strasbourg: Presses universitaires de Strasbourg.
Mazoyer, H. (2012) ‘La construction du rôle d’ingénieur-économiste au ministère
des Transports: Conseiller le politique, résister au comptable et discipliner le
technicien (1958–1966)’, Gouvernement et action publique 4(4), 21–43.
Mazoyer, H. (2013) ‘Réformer l’administration par le savoir économique: La
Rationalisation des choix budgétaires aux ministères de l’Équipement et des
Transports’, Genèses 93, 29–52.
Mazur, A. (2007) ‘Women’s policy agencies, women’s movements and a shifting
political context: Towards a gendered republic in France’, in J. Outshoorn,
J. Kantola (eds) Changing State Feminism, pp. 102–23, New York, Palgrave
Macmillan.
Morand, C.-A. (1999) Le droit néomoderne des politiques publiques, Paris: Librairie
Générale de Droit et de Jurisprudence (LGDJ).
Normand, R. (2011) Gouverner la réussite scolaire: Une arithmétique politique des
inégalités, Berne: Peter Lang.
Page, E. (2012) Policy without Politicians: Bureaucratic Influence in Comparative
Perspective, Oxford: Oxford University Press.
Pasquier, R. (2003) ‘La régionalisation française revisitée: Fédéralisme, mouvement
régional et élites modernisatrices (1950–1964)’, Revue française de science politique
53(1), 101–25.
Penissat, E. (2012) ‘Publier des “chiffres officiels” ou les contraintes bureaucratiques
et politiques qui façonnent l’expertise d’État: Le cas des statistiques du ministère
du Travail’, Gouvernement et action publique 4(4), 45–66.
Perret, B. (1999) ‘L’évaluation des politiques publiques dans les administrations
d’État, éléments de diagnostic’, Congrès de la Société française d’évaluation, Marseille,
4–5 June.
Pierru, F. (2007) Hypocrate malade de ses réformes, Bellecombe-en-Bauges: Croquant.
Pisani, E. (1956) ‘Administration de Gestion: Administration de Mission’, Revue
Quéré, O. (2014) L’Atelier de l’Etat. Des Instituts régionaux d’administration pour
former les cadres intermédiaires de la fonction publique (1966-2013), PhD Thesis,
Lyon: Université de Lyon: Institut d’Etudes Politiques de Lyon.
Quéré, O. (2015) ‘La fabrique des cadres intermédiaires de la fonction publique:
Enseigner et assigner un positionnement aux attachés dans les Instituts régionaux
d’administration’, Gouvernement et action publique 4(4), 33–54.
Quéré, O. (2017) ‘Construire l’État par son milieu: Les transformations du mandat
des cadres intermédiaires de l’administration’, Sociologie du travail, 59(3), http://
sdt.revues.org/828, DOI: 10.4000/sdt.828
Revillard, A. (2016), La cause des femmes dans l’Etat: Une comparaison France–Québec,
117
Rouban, L. (2010) ‘L’inspection générale des finances, 1958–2008: Pantouflage

et renouveau des stratégies élitaires’, Sociologies pratiques 21(2), 19–34.
Roubieu, O. (1994) ‘Le modèle du “manager”: L’imposition d’une figure légitime
parmi les hauts fonctionnaires des collectivités territoriales’, Politix 7(28), 35–48.
Roubieu, O. (1999) ‘Des managers très politiques: Les secrétaires généraux des
villes’, in V. Dubois, D. Dulong (eds) La question technocratique, pp. 217–31,
Strasbourg: Presses universitaires de Strasbourg.
Spenlehauer, V. (1999) ‘Intelligence gouvernementale et sciences sociales’, Politix,
12(48), 95–128.
Suleiman, E. (1979) Les élites en France: Grands corps et grandes écoles, Paris: Seuil.
Thoenig, J.-C. (1994) ‘Savoir savant et gestion locale’, Politix 7(28), 64–75.
Tirera, L. (2005) Du Commissariat général du plan au Centre d’analyse stratégique:
Stratégie de réforme d’une administration de mission, Paris: L’Harmattan.
Vauchez, A. (2008) ‘Le chiffre dans le “gouvernement” de la Justice’, Revue
française d’administration publique 125(1), 11–120.
Vigour, C. (2006) ‘Justice: L’introduction d’une rationalité managériale comme
euphémisation des enjeux politiques’, Droit et société 63–34(2), 425–55.
Viveret, P. (1989) L’évaluation des politiques et des actions publiques, Paris: La
documentation française.
118
SEVEN
Policy analysis in French

local government
Alain Faure and Emmanuel Négrier
The subject of the present chapter attempts to ‘deconstruct’ the French territorial
question and demonstrate that territorial policy analysis can break free from the
limitations imposed by the primarily ‘statist’ conceptual framework in which
it has hitherto been viewed. In the first section of the chapter, we indicate the
scale and nature of the present political and administrative territorial structures
and comment also on the paucity of academic research work on this subject. In
the second section, we highlight the main changes in the territorial framework
of policy building, through an evolution from vertical to horizontal dialectic of
powers and capacities. In the third section we discuss territorialisation, as the
results of a double process. On the one hand, the role of ideas in territorial policy
building; on the other hand, the dynamics of differentiation that put the French
model into question. Thus, focusing on policy analysis in French local policies –
in a global comparative perspective – sheds light on the huge challenge it causes
for the coherence of ‘national models’. As in other countries, local policies in the
French context are simultaneously influenced by globalisation and decentralisation,
opening new scenes for vertical and horizontal interactions toward the provision
of public goods and service.
An old, dense and underestimated territorial administration

The structures of French territorial government are notoriously complex.1 Many
of its contradictions and paradoxes can be traced back to its origins in the French
Revolutionary and Napoleonic period when the much-desired tabula rasa was put
into operation to destroy and even to eliminate the memory of the traditional,
often millennial, territorial structures. And yet, for all its rationalist features, the
new structures could scarcely avoid or conceal new and long-lasting conflicts,
concerning different meanings as well as applications of concepts such as ‘popular
sovereignty’, (‘direct democracy’ or versions of the ‘general will’), ‘federalism’
(compared with ‘Jacobinism’), plus conflicting theories of representativeness and
electoral practice, and, above all, the emergence of the ‘nation’ – as the paramount
unitary organisational structure which led Napoleon (particularly when faced with
foreign wars) to settle the arguments by streamlining the administrative system,
with an overwhelming stress on ‘top-down’ rather than ‘bottom-up’ processes.
For the following century and a half, this basic structure of French politics and
administration went largely unchallenged, at least until after the Second World
119
War when a new generation of young planners started to highlight the dangers of
Paris et le désert français. Motivated no doubt by a desire to resuscitate economically
and socially many of the run-down and neglected provinces as part of post-war
European regeneration, this strong desire for territorial regeneration became
a central thrust of government policy under the Fifth Republic, particularly
with the setting up of a devoted body: the Land Development and Regional
Action Delegation (Délégation à l’Aménagement du Territoire et à l’Action Régionale
[DATAR]) in 1963. In less than 30 years, this largely successful injection of
economic vitality into many parts of France had run its course, and, in the post-
Mitterrand years, emphasis has passed to a questioning of the more essentially
political functionality and efficacy of the complicated local and regional political
structures which have emerged (or survived or – worse still – thrived through
bureaucratic accretion), raising many questions concerning present policies and
the politics surrounding them.
The sheer density of French territorial administration is much greater than the
European average. In the three-tiered system of territorial government into which
the structure is normally divided, it is the basic unit of the ‘commune’ which
reveals the most startling diversity. With 36,767 communes, of which 32,000
have less than 2,000 inhabitants, France alone counts for more than 40 per cent
of the total number of ‘municipalities’ in the European Union. This makes the
use of this basic essential unit of limited value in functional normative terms. In
order to counter the obvious disparities created by this communal fragmentation,
French political forces have regularly encouraged intercommunal cooperation,
and this has led to the creation of a second local ‘unit’: intercommunalité. This level,
slowly but inexorably, has been ‘institutionalised’, given specific tasks, including
powers of taxation, with its own officials elected by universal suffrage, and its
own administration. This level of government now counts for 2,133 institutions.
Of these 11 are called metropoles (representing large urban conurbations) with
a further 238 ‘inter-communal communities’ in urban zones and 1,884 in rural
zones, some of which remain small (less than 5,000 inhabitants). Parallel to
these institutions which are called ‘Public Establishments for Intercommunal
Cooperation’ (EPCI), there are other bodies, without tax-raising powers,
whose role is to fulfil certain (often technical) tasks. These latter are called
‘intercommunal syndicates’ (numbering 13,402 in 2015) and it has been the desire
of successive governments since the 1990s to somehow ‘merge’ them into the
first intercommunal category (Kerrouche, 2012). We are not there yet!
Of the two ‘higher’ levels of territorial structures, the division of France into
‘départements’ was enacted in the first year of the Revolution (with the clear
intention of breaking the political and cultural power of the old provinces)
and Napoleon confirmed the departments’ ‘top-down’ function with the state
appointment of a ‘prefect’ to each department. The Fifth Republic has seen,
however, a major reduction in the powers of the prefect and, particularly in the
1980s, increasing power has been accorded to the departmental conseils which
mainly control local social policy. Threatened by developing institutions ‘above’
120
Policy analysis in French local government
and ‘below’ it, the département is the real phoenix of French administration,
often rising from the ashes because of strong support for it in the Senate and its
effective control over the implementation of social policy, redistributive aspects
of local solidarity and questions of transport (Négrier and Nicolas, 2011). It is
the département which is the crucial level at the intersection of urban and rural
zones. Its electoral system, direct universal suffrage at the canton level, tends to
reinforce the weight of rural zones and has recently been reformed, giving birth,
in 2015, to a unique two-round majority election of dual male/female candidates.
The modern ‘regions’, only date from 1964. While traditionally the weakest link
in French territorial administration (dating back no doubt to the revolutionaries’
fear of the power of the old provinces) the regions have progressively advanced in
recent years in political importance and agenda-setting policy-making (Barone,
2011). Their main thrust concerns economic development, education and higher
education and research, but also general territorial infrastructures. One very recent
development (2014–15) has been the (hasty) reduction in the number of regions
from 22 to 13 (excluding the five overseas regions), a ‘reform’ that reveals the
underlying paradox in attitudes to territorial government. Clearly intended to
save money by cutting out duplication of services at a time of general economic
retrenchment, the reform nonetheless officially aspires to create more powerful
regions, which can exist and thrive on a supra-national European level. To take one
example, the disparity between official aim and actual achievement is exemplified
in the amalgamation of Languedoc-Roussillon and Midi-Pyrénées in the south of
France. This new region is more than twice the size of Catalonia, its neighbour
in the north-east of Spain. But its consolidated budget comes to less than 10 per
cent of the Generalitat, its Catalonian institutional equivalent (Négrier, 2015).
This highlights the general problem – the relationship of the local and regional
resources (transfers from the central state plus limited tax-raising powers) to their
gradually increasing spheres of competence (resulting from the several waves of
decentralisation reforms particularly since the 1980s) and the corresponding
decline in direct central state territorial administration (Biland, 2012). It is true
that both the areas of competence and the resources have increased considerably,
to the extent that in 2012 local and regional services employed 1.8 million agents,
or 34 per cent of total public employment in France.
Table 7.1: Levels of territorial administration in France and associated expenditure

Collectivity Number Total expenditure 2012 (in €billions)
Communes 37,767 26
EPCI 2,133 39.58
Départements 100 71.35
Regions 26 27.92
Source: Personal extrapolation from distinct sources of Ministry of Interior figures, 2014
121
Increasing range of competences, growing human and financial resources,

enhanced legitimation through the application of universal suffrage – all these
would suggest a greater focus on territorialité in the overall political study and
analysis of French public policy. And yet strangely, the opposite seems to be the
case. A perusal of the content of the Revue Française de Science Politique in recent
years shows that any emphasis on the crucial importance of local and territorial
concerns in French policy studies remains a rarity. A similar analysis of the content
of the journal French Politics since 2010, which could serve as a barometer of
the extent to which the French territorial model is recognised internationally,
gives sparse results (one single article deals with decentralisation policy under
the Hollande presidency [Cole, 2014]) but without analysing the mechanisms or
other specificities of the French case; there are comparative studies integrating
France in general social policy (MacDaniel, 2014) or the evolution of the welfare
state (Simonet, 2014) but with no consideration of the territorial aspects of these
subjects.
Perhaps the reason for this neglect is the enduring prevalence of the historical
‘hierarchisation’ of French political values inferred above: ‘top-down’ has a
tendency to concentrate on the top. Localised nuances tend to get lost, and
simply obfuscate the international comparisons that emerge more clearly when
state models are not clouded by niceties of ‘local’ differences, which can easily
be relegated to what Freud dismissed as the ‘narcissism of minor differences’. But
if this marginalisation of the territorial issue applies to political science analyses
of French public policy and action (Douillet et al, 2012), French territoriality
nonetheless figures prominently in the English-speaking publications devoted
to urban, metropolitan and regional policies. An analysis of the content of the
International Journal of Urban and Regional Research, of the Journal of Urban Affairs
or of Regional and Federal Studies shows that France not only has its place in the
international comparisons made but also that the French ‘exceptionalism’ (often
implicit in the ‘state’ comparisons), is no longer so evident. Once we move away
from the national framework and put the focus on local configurations (regions,
metropoles, districts), then ‘national cultural factors’ fade into the background,
giving place to fruitful differentiations within the same national context (for
example: Lyon/Marseille, or Brittany/Ile-de-France) and (dispensing with
national borders entirely) facilitating very useful interterritorial comparisons
(Brittany/Galicia or Lyon/Barcelona).
Vertical and horizontal transformations

The historically ‘verticalist’ perspective, referred to above, needs further analysis,
before demonstrating the extent to which it has been superseded by increasing
‘horizontalist’ policy impulsions. ‘Horizontal’ does not imply that the horizons
are the very limited ones of localist autonomy but also include distant horizons,
thus giving a freer rein to international connections as well as international
comparisons, leading to what we propose as a new paradigm of territorialisation.
122
The ‘local’ is in any case a relative term. Mediation of ideas and practices can take
place at various levels with various horizons (particularly at the metropolitan level).
The verticalised ‘local’ and questioning the ‘top-down’ norm
The traditional vertical dialectic certainly dominated French territorial politics in

the early, Gaullian, years of the Fifth Republic. And the paradigms which were
current in the 1970s and 1980s, if trying to move away from the simple ‘top-down’
model, sought to identify points of equilibrium or of tension between the centre
and the periphery. Work on ‘régulation croisée’ (Grémion, 1976), in attempting to
locate such intersecting areas of tension, did indeed amount to a critique of the
normative discourse (often framed in juridical terms) which reduced territorial
politics simply to that of enacting state policy (Alliès, 1980). Nonetheless, local
powers were still essentially ‘heteronomous’, that is, dependent on ideological
and practical norms pertaining to the decision-making apparatus of the state. The
experiments carried out in certain major cities and other local projects were only
exceptions to the rule (Jobert and Sellier, 1977, 215). This ‘top-down’ vision was
also more or less shared by geographers, sociologists and economists who were
interested in questions of local government. Research on urban planning (Lojkine,
1972), on the city (Castells and Godard, 1974; Lefèbvre, 1973) or even on socio-
economic relations on a global scale, all tend to question any normative central/
local binary division – such a dichotomy is refuted as a false equality (Dulong
1978), with localism seen as a false identity (Sfez, 1977). Neo-marxist or other
overall systemic orders do not need to lay stress on the ‘local’.
Political science, however, focuses, on the tension at the territorial level between
the elective and the bureaucratic elements, leading to interesting variations from
any simple pattern, but all tending in the 1970s to be viewed in relation to a
‘central’ norm, the supposed coherence of which remained intact (Mabileau,
1972). Gradually, however, this was to change. Exceptions to the rule multiplied,
and with the French cumul des mandats (the holding by politicians of several local
or national mandates simultaneously), ‘power politics’ in some areas could no
longer be dismissed as small-time local politics of peripheral interest. There were
also cultural/political territorial manifestations (Lenclud, 1986) specific to certain
areas (Corsica, Languedoc, Brittany, for example). The analyses of these cases,
taking their differentiation from any national norm as a starting point, highlight
their socio-economic heterogeneity and the tenuous link of these to national
political integration (Jobert, 1977). Clearly, the top-down vertical and juridical
model had become an inadequate tool for satisfactory analysis.
The horizontal dialectic and international and interdisciplinary dimensions
It was at the beginning of the 1980s that the first major decentralisation movements
seriously called into question the adequacy of the vertical dialectic. This was also a
period when a generation of political scientists became more open to constructive
123
dialogue with other social scientists (particularly geographers, economists and

sociologists) in focusing on the territorial factor affecting social and political
issues. The ensuing controversies favoured the emergence of a more horizontal
conception of notions such as urban government (Borraz, 1998; Jouve and
Lefèvre, 1999), governance (John, 2001; Leresche, 2001), territorialised political
exchange (Négrier, 1998), the public/private relationship (Le Galès, 1997) and
also subsidiarity (Faure, 1998). These perspectives brought into play factors which
had been ignored in the top-down/Jacobin approach.
In the case of political sociology, in particular, such a reorientation of focus
is the key to a quite different approach to ascertaining what constitutes the
essence of localised politics (Briquet, 1997; Sawicki, 1997). Some of the work
follows up on older studies, in monograph form, of how politics itself and power
relations are constructed at the very basis of human behaviour (Lagroye, 1972, for
example). These types of analysis, are close to political anthropology, bringing
ethnology ‘back to French soil’ (Abélès, 1989; Pourcher, 1995). The local voice
is no longer just a sort of ‘state ventriloquist’ (Mabileau, 1993). Socio-historical
work on politics and administration reveal all that the bureaucratic mentality owes
to the emergence, in small towns, of a particular social group and its basic need
to establish institutional routines (Dumons et al, 1997). Other studies suggest
that public policy has developed, earlier and more effectively, policies that have
subsequently become an essential component of the new social contract. For
example, the decisive entry of women into local and regional governments, as well
as the reform of departmental elections (Troupel, 2017) contrasts with the erratic
implementation of national parity laws (Troupel, 2013). Apart from representation
issues, it was also at the local level that the first women’s rights policies were
forged, of which no one today disputes the centrality in contemporary democracy
(Mazur and McBride, 2006). It becomes very clear in this sort of work that
questions concerning the specificity of local space, legitimacy, social influence
and local circumstances and motivations are more important factors than whatever
regulations, procedures and party ideas are imposed by Paris.
In the case of the predominant ‘vertical’ dialectic, we observed that ‘top-down’
does not mean that all horizontal interactions were excluded (for example deals
made between the prefect and the influential ‘notables’ in a particular department).
In a similar way, the new emphasis on basic localism does not mean that other
variables are not also at play, factors such as economic competition, territorial
‘benchmarking’, Europeanisation, the weight of private interests – all these factors,
invoked in the 1990s, mean that we should not retain any simplistic romantic
vision of what local autonomy really means. The change of perspective rather
reflects a rejection of the post-Napoleonic juridical and rationalist organisational
dogmatism, and its replacement by a renewed (and largely international) interest in
institutional analysis (Hall, 1993; Hall and Taylor, 1996) as well as new reflections
on the meaning of ‘governance’ (Jessop, 1997; Marks, 1996) and on the workings
of an ‘urban régime’ (Harding, 1994). It should also be noted that regional analysis
is increasingly viewed in comparative terms (Keating and Loughlin, 1996; Jeffery,
124
1997; Le Galès and Lequesne, 1997; Négrier and Jouve, 1998). Metropolitan
questions have also moved away from a strictly national/top-down interpretation
towards focus on debates (reform vs public choice for example) which date back
a long time in English-language research (Woods, 1958; Ostrom et al, 1961).
These analytical models and concepts, such as ‘governance’, ‘urban regimes’,
‘new regionalism’, ‘new institutionalism’ – recently imported from English-
language publications – are not accepted wholesale as a coherent block without
a discriminating assessment of each term by the French academic community.
Urban governance is one of the clearest examples of the different variants of
meaning revealed (Jouve and Lefèvre, 1999; Lorrain, 2000; Gaudin, 2002; Le
Galès, 2003). Besides, other more continental imports have also had an impact
on French thinking about territorial politics. This is notably the case with the
Italian writings on industrial districts grouped around Carlo Trigilia (Trigilia,
1986; Ritaine, 1989; Benko and Lipietz, 1992) and on political exchanges
(Pizzorno, 1977; Ceri, 1981). These writings have also helped us to question
the hypothesis of the specific radical nature of the French case, and lead us to
consider French territoriality in comparative terms alongside other examples.
The whole territorial question is not about simply applying preconceived sets of
rules or central algorithms; rather, it opens up further questions, both empirical
and theoretical, about the meaning of public action and hence about the very
raison d’être of territorial policies (Arnaud et al, 2005).
An example. Lessons from the regionalisation of policy capacities

in three policy domains: education, transport, training
The results of the management of these areas by the regions is, in the opinion of
experts from each of them, especially noteworthy, given the initial capacity of the
regions. Not only the means that have been devoted to them have greatly exceeded
their investments consented before by the state through its central management,
but they have affected technicians, managers and elected officials who no longer
have any need to envy the professionals’ territorial administration of the state. This
professionalisation, which was initially inspired by the know-how developed by civil
servants, themselves transferred from the state, is no longer due to them: the regional
councils have become, especially in management positions and sector steering, particularly
attractive for more graduate and younger administrators (Bachelet, 2006). However,
this professionalisation has developed not against the state but in broad partnership
with it. This is what Thierry Berthet finds in the area of vocational training policies,
beyond the differences affecting, region by region, socio-economic structures of the
country, or even the way politicians do politics with such policies (Berthet, 2011).
The same assessment can be found for regional transport policies, for which Sylvain
Barone showed that once the regional professionalisation of these policies was observed,
these policies not only differed because of the topography of an area, but because of
former policies in the area, which necessarily constrained some of the choices ahead.
They also differed because of distinct forms of politicisation of this policy area throughout
different regions. To illustrate, in a region such as the Languedoc-Roussillon, Barone
125
notes that one vice president of the Regional Council from 2004 is not only a former
Communist Party leader, former Minister of transportation but also a former railwayman,
very sensitised to the maintenance of secondary lines and not just an apostle of speed
railway. Alongside more material factors, this has a clear influence on the regional policy
guidelines on the subject. In education, we draw similar lessons: a) a strong involvement of
institutions in quantitative terms, with a level of investment in the construction of school
facilities far higher than what the state spent in it, and at a faster pace; b) professional
management in partnership with state that, in this area, has retained the management
of personal and educational programmes; c) a relative differentiation of educational
policies across regions, less according to a global ideology (left against right politics
and policies) than the way politicians considered educational policies within their own
prospects of political legitimacy (Dupuy, 2011). We can take the example of some policy
tools implemented during a period – between 2004 and 2014 – where almost all regions
were managed by French left governments. A regulation indeed considered as ‘leftist’
(the provision of school books to children according with resource conditions, that is,
positive discrimination) was often preferred while it was rather ‘rightist’ in its political
orientation. Similarly, if early in their management of academic competence, the left-wing
regional governments were more reserved than those belonging to the right wing about
private education funding, this ideological ‘marker’ has almost disappeared over time.
Basically, the regionalisation experience of public policies led to some paradoxes:

implemented by politicians, it has largely depoliticised; in sectors full of ideology
and collective imagination, ideas into action have defused rather than confirmed
hopes or fears; expertise, less autonomous than shared with the authorities of
different levels, here leads to disappointment: the ideas it produces dissolve the
ideological content of public policies.
The emergence of a new paradigm of territorialisation

At the heart of these questions, there lies a question of scale. The horizontal
against the vertical has to be re-examined in the light of the changing nature
and the growing dimensions of territorial politics. New ideas are permeating
the new territorialisation processes – more levels of expertise, suitable to each
different local context.
A political inversion
The clearly discernible movement towards both further decentralisation and

also towards Europeanisation has led to an inversion of tendencies, with a
repoliticisation of the state and a depoliticisation and professionalisation of
territoriality. There has been a change of scale, both political, administrative
and spatial, with a move to transfer a great number of public services to the
intercommunal level. At the same time, the relations between the different scales
have also undergone transformation (Faure et al, 2007). The territorial authorities
126
now take on a growing proportion of what is still called ‘welfare state’ activities,
now including social, educational, housing and cultural policies, with the state
playing a more distant role, often resorting to contractualisation (Epstein, 2004)
and placing greater reliance on market forces rather than directly attributing
responsibility for, and exercising direct control over, specific activities. The result
of this is that local elites are more professionalised and better equipped (and with
more funding). The idea of the state as master of public policy is replaced by
local governments, with their new attributions, developing their own expertise,
taking more management decisions as well as more control over policy (Borraz
and Négrier, 2007). We should note that this inversion means that the state
cannot remain aloof from or be impervious to local politics/policies (as it did in
the early Fifth Republic) and this brings us back to the more fluid interactions
between central and local government during the Third and Fourth Republics
(Le Lidec, 2001).
Territorial ideas in action
Does this shifting of the centre of gravity of public policy imply new constraints
and a change in the set of paradigms pertaining to the discourse of territorial
government? This is the subject of much discussion between scholars. Local
districts cannot themselves arrogate, unilaterally, the expression of their own
political ‘world’. Ideas have to be more generalised in their origin than that and
the border between the public and private influences is difficult to draw. Taking the
‘urban’ question as an example, the word of the state professionals on this subject
is now diluted by the more diffuse and variegated expertise expressed by local
public bureaucracies, by circles of expertise emerging from the mixed economy,
from consultancy agencies and international bodies (Cadiou, 2005). It is true
that the essential plurality of this new world will not prevent the emergence of
dominant discourses. This plurality is clearly one of the most significant current
territorial changes. For example, one observes a certain standardisation in the
production of ideas on ‘urban projects’ (Pinson, 2005), but the acceptance of this
standardisation remains relatively limited, given the strong variation in argument
from one locality to another and the important restructuring changes that are
underway.
There were of course in the past distinctive political ‘visions’ which were
rooted in certain areas of France. One thinks of ‘municipal socialism’ which
was a powerful vector of partisan identification of public policy in certain areas
(Lefebvre 2006). One thinks also of other ‘markers’ of leftwing municipal policy,
such as ‘cultural action’ (Négrier, 2003) or ‘social housing’ policies (Maurin,
2004; Donzelot, 2006). In a certain way, it is paradoxically the decline of these
‘great ideals’ which has opened the way for the new expression of territorial
action. This turning point of institutional reform of urban government can be
illustrated by analysing the two forms of public action, diametrically opposed,
which structure current local government debate along the lines of reform
127
versus public choice. The first word (reform) implies tackling the new problems
of urbanisation by creating new institutions which, in their scope and potential
targets, will adequately deal with the new scale of urban problems (Wood, 1958).
The problem here is that no two urban conglomerations are remotely the same
(Aix/Marseille is fundamentally different from a ‘Greater Paris’ or a ‘Greater
Grenoble’). Whereas the second phrase (public choice) implies that institutional
fragmentation, entailing little new bureaucratic structures, leaves the field open
to fiscal competition and competing services between communes, intending
thereby to improve the efficacy of local government through ‘customer/residents’
choice (Ostrom et al, 1961).
What lessons can be learnt from these dilemmas? Above all that ideas must
take concrete form at the urban and metropolitan levels themselves. Ideas only
receive lukewarm endorsement when they have been thought up elsewhere. Ideas
originating in the very place of their implementation will stand a better chance of
being well and truly implemented. The setting up of an urban conglomeration,
for example, has to face directly concrete problems of specific kinds: the type of
representation for participating communes in the new cooperative body (Le Saout
and Madoré, 2004); the effectiveness of their public policies (Négrier, 2007);
fairer ways of tax redistribution (Rousseau, 2004) and above all the problem
of justifying (through a new public discourse) the existence of a new level of
territorial government by a clear demonstration of the need for sharing (public
transport costs, school provision, social housing, cultural facilities, and so on).
Even the virtually new word ‘métropole’ impedes, by its recent – and sometimes
ambiguous – meaning, the development of a discourse which can be impassioned
and convincing.
The variables of local differentiation
The territory has ceased to be ‘heteronomous’, to become a sphere in which

political interests, the differing logics of sectors, professional expertise and civil
society’s expectations, come together in a unique combination. This dynamic
of differentiation is as far removed as possible from what remains of the ‘Jacobin’
(or Napoleonic) requirement for a unified national ‘territory’, designed like a
classical French garden.
When having to arbitrate priorities, the regional, departmental and metropolitan
administrations can no longer treat small differences as narcissistic distractions
from the big picture, as in the early days of decentralisation. Local objectives and
satisfaction are the primary concern. Dossiers are no longer easily categorised in
nationally unified rational terms and multiple types of expertise must constantly be
tested, evaluated and re-evaluated. This new form of public action, that has to be
reflected also in national and international regulations, has recently been likened
by the political scientist Pierre Muller to the coming together of three tectonic
plates, a metaphor (albeit a potentially cataclysmic one) for the interaction of
the play of market forces, the exigencies of the ‘public space’ and the traditional
128
dirigisme (even if much reduced) of the state (Muller, 2015). These are the forces
to be reconciled in each large urban area.
Two variables must constantly be taken into account in reconciling these three
factors in our way of studying territories. The first is of a cultural socio-historical
nature and concerns the weight of a location’s specific past, that is, the depth,
values and circumstances which condition the socio-cultural proclivities and the
choice of options of the actors concerned. Comparative studies show that power,
authority, institutions and democracy are not thought of in the same way, even
if the stakes are the same, depending on whether one is recounting the common
good in Madrid, Barcelona, Naples, Milan, Marseille, Nantes, Frankfurt or
Munich. Each urban region has its own ‘dependency path’ of institutions and
values which produce quite different public actions, even within the same national
space. The second variable is more materialist and is also topographic. Depending
on the location, for example in a mountainous or maritime area, in a dense or
relatively sparsely populated area, the territorial location does not point to the
utilisation of the same political instruments.
For the analyst, whether as an engaged expert or a distanced observer, taking
these two variables into account poses complicated problems of method and
conceptualisation. In the nascent French research devoted in recent years to
urban and regional governments, these imponderables have been translated into
a series of new challenges: opening up new frontiers for public action (Mevellec,
2008), leading to irreversible changes in our conception of political power (Ben
Mabrouk, 2006), creating new discursive narratives to fit the metropolitan pattern
(Tomàs, 2012), assessing the impact of the rise of regional governments on social
fragmentation (Dupuy, 2011), and reflecting on how local elected officials are
invested with new missions of political mediation (Faure, 2016). In France, as
elsewhere, the rise in power of metropolitan areas and of regions suggests the
emergence of a democratic differential in the sense that there will be in future
powerful territorial ways of producing order and enunciating meaning. This
new equation is thrust into the very heart of the research agenda in political
science and suggests bridges to connect several interpretations and conceptual
frameworks hitherto considered as competing and irreconcilable. To the brave
accomplishments of ‘new public management’ or of ‘public choice’ can now be
superposed a whole motley of hypotheses which echo the analytical innovations
sometimes qualified as turning points (territorial, narrative, linguistic, emotional,
participative and so on). The perspectives opened up by all this can make one
a little giddy, for they link the ‘big questions’ concerning power and political
legitimacy to our ability to see these questions through new scientific paradigms.
Conclusion
Local government and territorial policies have changed over the past 30 years.
They contribute to transforming political analysis itself, suggesting new objects,
but also new epistemological challenges. Thus, the three perspectives we’ve
129
highlighted (professionalisation, pluralisation, differentiation) offer social scientists

unprecedented challenges in France, although often better established in other
European countries.
The first perspective – the professionalisation of local policies – requires analysts
to produce models more aware of territorial configurations. It leads to paying
more attention to ground research (less theoretical, more empirical) and to new
links with other scientific disciplines (geography, history, law, anthropology).
The second perspective – the pluralisation of actors – implies a renewed work
on local narratives and discourses. It would be tempting to only consider the
functional issues of multi-layered management, as generally understood in terms
of overcrowded policy making. But territorial governance challenge is less technical
than symbolic and semantic. This pluralisation means at first the crisis of the great
and abstract narrative on general interest (Faure, 2007): a double crisis both in
terms of membership and in terms of efficiency in the everyday political life of
cities and regions.
The third perspective – territorial differentiation – leads to a double and
especially critical issue in France. The first is political and economical: producing
knowledge on the conditions of a ‘positive differentiation’ (Négrier, 2010) that
combines fiscal, legal and philosophical prospects. Here, the surrender of the
territorial equality myth is only the starting point of a huge project: the invention
of territorial equity. The second issue questions local identity. Behind this term,
extremely opposed conceptions of social and spatial diversity can be found.
Between territorial identity evoked in active strategies of diversity and that of
essentialist neo-regionalism, there is a tremendous chasm. The Catalan experience,
as well as those of Lombardy and Scotland, or the issues related to migration in
contemporary Europe show that these controversies must be explored. Here, the
role of experts is fundamental: experts who invite us to understand all claims of
identification but to be wary of any pretension to identity.
These three perspectives highlight the extent to which local policy analysis is
now confronted by a crucial issue: to go beyond its ‘natural’ tendency to produce
specialised diagnostics and develop an original perspective about a political/policy
model of democracy both influenced by standardisation and differentiation. This
double dialectics (politics/policy; standard/difference) has more to offer than the
ritual opposition between institutionnalisms and culturalisms. And the French case,
much more acquainted with standards, national institutions and politics, should
be the country for assessing differences, regional and policy contrasted influences.
Note
1 The French adjective territorial, in an administrative sense, like its derivative territorialité, tends
to cover all tiers of government up to but not including the central state.
130
References
Abélès, M. (1989) Jours tranquilles en 1989, Paris: Odile Jacob.
Alliès, P. (1980) L’invention du territoire, Fontaine: Presses universitaires de Grenoble.
Amin, S. (1973) Le développement inégal: Essai sur les formations sociales du capitalisme
périphérique, Paris: Éditions de Minuit.
Arnaud, L., Le Bart, C., Pasquier, R. (2005) ‘Déplacements idéologiques et action
publique: Le laboratoire des politiques territoriales’, Sciences de la société 65, 3–7.
Bachelet, F. (2006) ‘Sociologie, formation et carrière des hauts fonctionnaires
territoriaux’, Annuaire des collectivités locales 26, 99–113.
Barone, S. (2011) Les politiques régionales en France, Recherches Politiques, Paris:
La Découverte.
Ben Mabrouk, T. (2006) Le pouvoir d’agglomération en France: Logiques d’émergence
et modes de fonctionnement, Paris: L’Harmattan.
Benko, G., Lipietz, A. (1992) Les régions qui gagnent. Districts et réseaux: les nouveaux
paradigmes de la géographie économique, Paris: Presses universitaires de France.
Berthet, T. (2011) ‘La régionalisation de la formation professionnelle’, in S.
Barone (ed) Les politiques régionales en France, pp. 51–64, Recherches Politiques,
Biland, É. (2012) La fonction publique territoriale, Repères/Sciences Politiques-
Droit, Paris: La Découverte.
Borraz, O. (1998) Gouverner une ville: Besançon,1959–1989, Rennes: Presses
Borraz, O., Négrier, E. (2007) ‘The end of French Mayors?’, in J. Garrard (ed)
Heads of the Local State in Past and Present, Basingstoke: Palgrave Macmillan.
Briquet, J.-L. (1997) La tradition en movement: Clientélisme et politique en Corse,
Paris: Belin.
Cadiou, S. (2005) ‘Vers une action urbaine “moderniste”: Les effets du discours
des experts savants’, Sciences de la société 65, 9–27.
Castells, M., Godard, F. (1974) Monopolville. L’entreprise, l’État, l’urbain, Paris:
Mouton.
Ceri, P. (1981) ‘Le condizioni dello scambio politico’, Quaderni di Sociologia 4,
640–63.
Cole, A. (2014) ‘Not saying, not doing: Convergences, contingencies and causal
mechanisms of state reform and decentralisation in Hollande’s France’, French
Politics 12(2), 104–35.
Donzelot, J. (2006) Quand la ville se défait, Paris: Le Seuil.
Douillet, A.-C., Faure, A., Halpern, C., Leresche, J.-P. (2012) L’Action publique
locale dans tous ses États: Différenciation et standardisation, Paris: L’Harmattan.
Dulong, R. (1978) Les régions, l’Etat et la société civile, Politiques, Paris: Presses
Dumons, B., Pollet, G., Saunier, P.-Y. (1997) Les élites municipales sous la IIIème
République: Des villes du Sud-Est de la France, Paris: Centre national de la recherche
scientifique (CNRS).
131
Dupuy, C. (2010) Politiques publiques, territoires et inégalités: Les politiques régionales

d’éducation en France et en Allemagne (1969–2004), PhD thesis, Paris: IEP de Paris.
Dupuy, C. (2011) ‘Y a-t-il de la politique dans les politiques régionales
d’éducation?’, in S. Barone (2011) Les politiques régionales en France, pp. 65–84,
Epstein, R. (2004) ‘Après la territorialisation, la différenciation territoriale?’,
Pouvoirs Locaux 63(4), 407–33.
Faure, A. (ed) (1998) Territoires et subsidiarité, Logiques Politiques, Paris:
L’Harmattan.
Faure, A. (2007) ‘Une nouvelle critique territoriale?’, in A. Faure, E. Négrier (ed)
Les politiques publiques à l’épreuve de l’action locale, pp. 275–83, Paris: L’Harmattan.
Faure, A. (2016) Des élus sur le divan. Les passions cachées du pouvoir local, Grenoble:
Presses Universitaires de Rennes.
Faure, A., Leresche J.-P., Muller P., Nahrath S. (eds) (2007) Action publique et
changements d’échelles: Les nouvelles focales du politique, Paris: L’Harmattan.
Garraud, P. (1983) ‘Le recrutement des maires en milieu urbain: Esquisse de
typologie’ Pouvoirs 28, 29–44.
Gaudin, J.-P. (2002) Pourquoi la gouvernance?, La bibliothèque du citoyen, Paris:
Presses de Science-Po.
Ghorra-Gobin, C. (2004) ‘L’étalement de la ville américaine: Quelles réponses
politiques?’, Esprit 3–4, 145–59.
Grémion, P. (1976) Le pouvoir périphérique: Bureaucrates et notables dans le système
politique français, Paris: Seuil.
Hall, P. (1993) ‘Policy paradigms, social learning and the state’, Comparative
Politics 25(3), 275–96.
Hall, P.A., Taylor, R.C.R. (1996) ‘Political science and the three new
institutionalisms’, Political studies 44(5), 936–57.
Harding, A. (1994) ‘Urban regimes and growth machines: Towards a cross-national
research agenda’, Urban Affairs Quarterly 29(3), 356–82.
Jeffery, C. (1997) The Regional Dimension of the European Union: Towards a Third
Level in Europe?, London: Frank Cass.
Jessop, B. (1997) ‘Capitalism and its future: Remarks on regulation, government
and governance’, Review of International Political Economy 4(3), 561–81.
Jobert, B. (1977) ‘Bureaucraties sociales et planification locale’, Annuaire de
l’Aménagement du territoire, vol 8, pp. 237–50, Fontaine: Presses universitaires
de Grenoble.
Jobert, B., Sellier, M. (1977) ‘Les grandes villes: Autonomie locale et innovation
politique’, Revue Française de Science Politique 27(2), 205–27.
John, P. (2001) Local Governance in Western Europe, London: Sage.
Jouve, B., Lefèvre, C. (1999) ‘De la gouvernance urbaine au gouvernement des
villes? Permanence et recomposition des cadres de l’action publique en Europe’,
Revue Française de Science Politique 49(6), 835–53.
132
Jouve, B., Négrier, E. (2008) ‘La sécession métropolitaine: Analyse comparée

franco-américaine’, in M. Camau, G. Massardier (eds) Les régimes politiques
revisités, Paris: Khartala.
Keating, M., Loughlin J. (1996) The Political Economy of Regionalism, London:
Frank Cass.
Keil, R. (2000) ‘Governance restructuring in Los Angeles and Toronto:
Amalgation or secession?’, International Journal of Urban and Regional Research
24(4), 758–81.
Kerrouche, E. (2012) ‘Bilan de l’intercommunalité à la française’, Revue
française d’administration publique 141, 37–53.
Lagroye, J. (1972) Société et politique. Jacques Chaban-Delmas à Bordeaux, Paris:
Pédone.
Le Galès, P. (1997) ‘Quels intérêts privés pour les villes européennes?’ in A.
Bagnasco, P. Le Galès (eds) Villes en Europe, pp. 231–54, Paris: La Découverte.
Le Galès, P. (2003) Le retour des villes européennes, Paris: Presses de Science Po.
Le Galès, P., Lequesne, C. (1997) Les paradoxes des régions en Europe, Paris: La
Découverte.
Le Lidec, P. (2001) Les maires dans la République: L’association des maires de France,
élément constitutif des régimes politiques français depuis 1907, thesis, Paris: Université
de Paris 1, Panthéon-Sorbonne.
Le Saout, R., Madoré, F. (2004) Les effets de l’intercommunalité, Rennes: Presses
Lefèbvre, H. (1973) Espace et politique, Paris: Anthropos.
Lenclud, G. (1986) ‘De bas en haut, de haut en bas, le système des clans en Corse’,
Etudes Rurales 101(1),137–73.
Leresche, J.-P. (2001) ‘Gouvernance et coordination des politiques publiques’,
in J.-P. Leresche (ed) Gouvernance locale, cooperation et légitimité: Le cas suisse dans
une perspective comparée, pp. 29–65, Paris: Pédone.
Lojkine, J. (1972) La politique urbaine dans la région parisienne, Paris: Mouton.
Lorrain, D. (2000) ‘Gouverner les villes: Questions pour un agenda de recherche’,
Pôle Sud 13, 27–40.
Lowery, D. (2001) ‘Metropolitan governance structures from a neoprogressive
perspective’, Swiss Political Science Revue 7–3, 130–6.
Mabileau, A. (1972) Les facteurs locaux de la vie politique nationale, Paris: Pedone.
Mabileau, A. (1993) A la recherche du ‘local’, Paris: L’Harmattan.
MacDaniel, S. (2014) ‘Post-crisis social democratic policy capacity in France and
the United Kingdom: A lesson from the globalisation and social democracy
debate’, French Politics 12(4), 283–309.
Marks, G. (1996) ‘European integration from the 1980’s: State-centric v multi-
level governance’, Journal of Common Market Studies 34(3), 341–78.
Maurin, E. (2004) Le ghetto français. Enquête sur le séparatisme social, Paris: Seuil.
Mazur, A., McBride, D. (2006) ‘The RNGS data set: Women’s policy agencies,
women’s movements and policy debates in western post-industrial democracies’,
French politics 4(2), 209–36.
133
Mevellec, A. (2008) La Construction politique des agglomérations au Québec et en

France, Laval: Presses de l’Université de Laval.
Muller, P. (2015) ‘Une théorie des cycles d’action publique pour penser le
changement systémique’, in L. Boussaguet, S. Jacquo, P. Ravine (eds) Une
de Sciences Po.
Négrier, E. (1998) ‘Échange politique territorialisé et intégration européenne’, in
R. Balme, A. Faure, A. Mabileau (eds) Les nouvelles politiques locales: Dynamiques
de l’action publique, pp. 111–34, Paris: Presses de Science Po.
Négrier, E. (2007) ‘Penser les changements d’échelle territorial: Institution,
dynamiques sociales et politiques métropolitaines’, in A. Faure, J.-P. Leresche,
P. Muller, S. Nahrath (eds) L’action publique à l’épreuve des changements d’échelle,
Négrier, E. (2010) ‘La différenciation positive’, in J.C. Nemery (ed) Quelle
nouvelle réforme pour les collectivités territoriales françaises?, pp. 315–24, Grale, Paris:
L’Harmattan.
Négrier, E. (2015) ‘Réforme territoriale: Le pire n’est pas sûr’, NectArt (Nouveaux
Enjeux sur la Culture, les Transformations artistiques et la revolution
Technologique) 1(1), http://www.nectart-revue.fr/1-negrier/
Négrier, E., Jouve, B. (1998) Que gouvernent les régions d’Europe?, Logiques
Politiques, Paris: L’Harmattan.
Négrier, E., Nicolas, F. (2011) ‘Cumbersome inheritance or democratic lever:
The department at the center of French territorial concerns’, in H. Heinel, X.
Bertrana (eds) The Second Tier of Local Government in Europe: Provinces, Counties,
Départements and Landkreise in comparison, pp. 73–89, London: Routledge.
Ostrom, V., Tiebout, C., Warren, R. (1961) ‘The organisation of government
in metropolitan areas: A theoretical inquiry’, American Political Science Review
55(4), 831–42.
Pinson, G. (2005) ‘l’idéologie des projets urbains’, Sciences de la société 65, 29–51.
Pizzorno, A. (1977) ‘Scambio politico e identità collettiva nel conflitto di classe’, in
C. Crouch, A. Pizzorno (eds) Conflitti in Europa, pp. 407–33, Milan: Etas Libris.
Pourcher, Y. (1995) Les maîtres de granit: Les notables de Lozère du XVIIIe siècle à
nos jours, Paris: Plon.
Ritaine, E. (1989) ‘La modernité localisée? Leçons italiennes sur le développement
regional’, Revue Française de Science Politique 39(2), 154–78.
Rousseau, M.P. (2004) La nouvelle architecture fiscale locale et les enjeux d’intégration
et de développement économique des territoires, Rapport ŒIL, pour le Ministère de
l’Intérieur, Paris: Université Paris XII.
Sawicki, F. (1997) Les réseaux du parti socialiste. Sociologie d’un milieu partisan, Paris:
Belin.
Sfez, L. (1977) L’objet local, Paris: Christian Bourgois.
Simonet, D. (2014) ‘The convergence of the British and French models of
capitalism’, French Politics 12(1), 1–21.
134
Tomàs, M. (2012) Penser métropolitain ? La bataille politique du Grand Montréal,

Québec: Presses de l’Université du Québec.
Trigilia, C. (1986) Grandi partiti e piccole imprese: Comunisti e democristiani nelle
regioni a economia diffusa, Bologne: Il Mulino.
Troupel, A. (2013) ‘Entre consolidation et remise en cause: Les tribulations de
la loi sur la parité (2000–2010)’, Modern and Contemporary France 21(February),
17–36.
Troupel, A. (2017) ‘Élus’, in N. Kada, R. Pasquier, C. Courtecuisse, V. Aubelle
(eds) Dictionnaire encyclopédique de la décentralisation, pp. 483–9, Paris: Berger-
Levrault.
Wood, R. (1958) ‘The New Metropolis: green belts, grass roots or Gargantua’,
American Political Science Review 52, 108–22.
135
EIGHT
Beyond weakness: policy analysis in the

French parliament
Olivier Rozenberg and Yves Surel
Not all parliamentary activities call for policy analysis. Some, such as oral questions,
require less expertise and a good share of bad faith. Others are based on log-rolls
and on strategic or signalling games. To some extent, however, Members of
Parliament (MPs) in modern legislatures are expected to care about the content of
public policies. Despite a shared trend toward de-parliamentarisation (Rosanvallon,
2015), most legislation still has to be formally approved by parliament in most
democracies. Beyond passing laws, parliaments can also oversee government
activities and public policies. The electoral and institutional stakes of such oversight
activities are high since, all over Europe, including in France, legislatures can decide
to dismiss governments not only for legal reasons but also for political reasons.
Any legislature actually finds itself in a twofold situation vis-à-vis policy analysis.
First, it relies on policy analysis produced beyond its walls to perform its main
tasks: legislating, overseeing the government and participating in public debates.
In order to do so, legislatures in Europe crucially depend on external expertise
because their human resources are limited. The topics they are supposed to
cover are potentially unlimited. Moreover, in most cases, MPs have to act in
emergency situations. Parliaments, however, are also loci where policy analysis
is produced through a variety of procedures: inquiry committees, reports of all
kinds, answers obtained from the administration in response to oral or written
questions for example, therefore producing reliable information about public
policies can be either a by-product of parliamentary activities (as is the case for
committee reports published during the legislative process) or an end in itself (as
for inquiry committees’ reports).
This chapter assesses the role of the French parliament which comprises two
assemblies, the National Assembly (NA) and the Senate, based on this twofold
dimension (using and producing expertise). Should the shared and well-known
diagnosis of the weakness of the French parliament under the Fifth Republic
(Huber, 1996; Hayward, 2004; François, 2007) be extended to the policy analysis
dimension? This may seem logical. Indeed, given that the process of expertise is
a key dimension of actual legislative games (Krehbiel, 1991), one might expect
the French government to strictly bind the houses’ capacity to use and produce
information. Yet, the alternative hypothesis can also be formulated: the French
parliament might have sought to offset its constitutional boundaries by relying
and investing in expertise. Collecting reliable information or producing relevant
public policy assessments would thus constitute a sort of soft power for the
137
backbenchers’ ends. In this chapter, both views are tested. First, the numerous
limits encountered by the French parliament are analysed. Second, the chapter
examines MPs’ capacity to initiate and use policy analysis.
A parliament in a weak position

The limitations of the French parliament’s capacity regarding policy analysis are
numerous. They are rooted in the practice and spirit of the regime despite an
ambitious reform of the Constitution in 2008. They also result from the type of
human resources available in each assembly and from how these human resources
are managed.
A bounded parliamentarism
There is a rare consensus among scholars and practitioners about the weakness of
the French parliament. Indeed, Philip Norton (1990) once wondered whether
the French parliament could even be defined as a legislature. The significance
of the 1958 shift largely explains this consensus. Before 1958, under the Fourth
Republic, France was characterised by a traditional parliamentary regime where
unstable parliamentary majorities regularly broke governments. In 1958, De Gaulle
famously conditioned his come-back on a new constitutional settlement that was
far less favourable to parliament. Constitutionally, the legislative branch faces two
types of disadvantages. First, the President of the Republic tends to be the real
leader of the executive power to the detriment of the Prime Minister. Since 1962,
this President is elected not by parliament but directly by the people and, most
important, parliament can neither dismiss nor oversee him. This constitutes a
major source of disequilibrium because the NA (but not the Senate) can be easily
dissolved by the President. Second, there is still a government which is headed
by a Prime Minister and which can be censured by the NA (not the Senate), but
many constitutional provisions are detrimental to the parliament. Those rules
are numerous and cover all aspects of political life: government formation and
resignation, the oversight of the government, the definition of the legal field and,
last but not least, the legislative procedure.
In addition to these numerous constitutional limitations, a political limitation
was added in 1962 with the election of a pro-President majority in the NA
following a tense dissolution. Ever since, presidents have usually had at their
disposal a rather solid majority in the NA which strictly follows the voting
instructions of the government. Many elements contribute to maintaining what
has been called le fait majoritaire: the choice of the voters, the threat of dissolution,
the effect of the plurality electoral system and the ascendancy of the presidential
election over political life. Exceptions to this have been the three periods of
divided governments (cohabitation) in 1986–88, 1993–95 and then the longest
one in 1997–2002. As a result of the opposition between the NA majority and
the President, the Prime Minister was de facto imposed on the President and held
138
Beyond weakness
most of the power. However, two reforms undertaken in the early 2000s have
made such a situation less likely: the alignment of the duration of the presidential
term with the legislative one (five years) and the decision to vote for the President
before the NA – the parliamentary majority to some extent owing its victory
to the President.
As a result of this constitutional imbalance, the capacity of the Parliament to
initiate bills or to define autonomously its own agenda is strictly bounded (Gicquel
and Gicquel, 2015). An aspect of the asymmetry between the executive and
legislative power can be observed through the origin of adopted bills. Since the
last legislative elections in 2012, 372 texts have been presented by the government
whereas 1,517 texts have been proposed by MPs. However, if we consider the
texts that were ultimately adopted, 277 texts proposed by the government were
enacted (75 per cent) as opposed to only 82 private bills (25 per cent).1 One
quarter of the law production is certainly not negligible but an insight into the
issues covered reveals that private bills often tend to focus on secondary topics
related to day-to-day life rather than on major society choices. In addition,
some of the private bills indirectly originate from the government because some
ministers seek to avoid endorsing them publicly.
Moreover, the ability of the parliament to develop an autonomous expertise on
policies is further constrained by limitations on its capacity to amend the texts it
has initiated. While the Constitution (article 44) states that both the government
and MPs have the right to amend legislative proposals, some additional dispositions
privilege the government. For example, in the plenary session, the government
can oppose an amendment which has not previously been examined by the
relevant legislative committee. Article 44-3 also provides for a procedure for
blocked votes which authorises the government to ask for a vote on a text with
the amendments it has presented or accepted. As a result of this set of restrictive
rules, during the current term more than 100,000 floor amendments, presented
primarily by MPs, have been proposed, but only 15,650 of them have been
adopted (15 per cent for 2012–15). Article 49-3 of the Constitution also sets
out another famous specific constitutional provision. It allows the government
to pass a bill without voting except in circumstances where its opponents are
ready to censure the government. This kind of institutional blackmail, which
makes it possible to clarify who is in the majority and who is not (Huber, 1996),
has been used essentially when the majority is uncertain (1976–81) or reduced
(1988–93). More recently, it was activated by President Hollande to circumscribe
the rebellion among the parliamentary troops.
These data and new developments are consistent with previous observations in
comparative research on political regimes and/or on parliaments (see for example
Döring, 1995; Norton, 2002; Sieberer, 2011). The French parliament remains
one of the weakest legislative institutions among contemporary political regimes
when it comes to the assessment of its role in the decision-making process. As a
consequence, it is fair to say that there are limited incentives for the development
of a real expertise in policy analysis within the parliament. Since the influence of
139
MPs on policies is weak, they are less inclined to specialise in policy issues and to
develop any expertise in several policy domains. This is all the more true given
that parliament is composed of a limited number of standing committees: eight
in the NA since 2009 for 577 députés. Compared to the US Senate, where there
are 16 standing committees and 88 committees for 100 senators, French MPs
are not called upon to specialise and work actively on specific issues with their
colleagues and with the relevant private actors. In many instances, debates are
therefore more political than technical and they are rarely based on an autonomous
policy analysis.
From a policy analysis perspective, the constitutional leadership of the President
is a source of difficulty for the assemblies. The problem is not just that the leader
de facto initiating most of the bills cannot be censured in parliament; the greatest
problem is that the President cannot be heard or questioned in each assembly. The
President is indeed not accountable to the Parliament. Since 2008, although the
President has been able to deliver a speech to both assemblies jointly in Versailles,
the Constitution specifies that the President should leave the hemicycle should
a debate take place afterwards.
In addition to the constitutional text, the interpretation by the Constitutional
Council since the very beginning of the Fifth Republic contributed to limiting
the capacity of parliament to develop several kinds of non-legislative activities,
including policy evaluation. In a famous 1959 decision, the Council stressed that
the NA could only control the government through the procedures that were
specifically provided for in the Constitution, that is, confidence votes.2 The
justified desire to preserve the stability of the government contributed, in the
long run, to establishing less justified restrictions on the oversight capacity of
the parliament vis-à-vis the government. To give just one example, the Council
declared in 2009 that the debates of a new NA evaluation committee could
not be made public as it was the task of the whole assembly, and not of a given
committee, to control the government.3 More generally, this interpretation led
to the emergence, in the legal doctrine, of a distinction between the oversight of
the government, a political control that could virtually lead to censure, and the
assessment of administration and public policies, a more neutral control lacking
coercive legal means. We will return to the damaging effect of this abstract
distinction in terms of political mobilisation in our conclusion.
The 2008 Constitutional revision: a useless reform?
Elected in 2007, the right-wing President N. Sarkozy launched a major

constitutional reform officially aimed at strengthening the French parliament
(Rozenberg, 2016; Pouvoirs, 2013). Although there were other motivations such
as the willingness to limit parliamentary filibustering, and although Sarkozy did
not want to depart from the so-called ‘logic of the Fifth Republic’, the project
was undoubtedly ambitious. It was all the more ambitious as it was delegated
to academics and politicians outside the administration. These constitutional
140
Beyond weakness
entrepreneurs were seemingly convinced that the key to legislative influence was
information and expertise, and that parliament should not only legislate but should
also assess legislation. Many of the amendments passed were thus connected to
the policy analysis capacity of parliament.
The first change was symbolic and concerned the definition of parliament.
In addition to the traditional functions of legislation and oversight, article 24
of the Constitution now stipulates that parliament ‘shall assess public policies’.
This provision paved the way for the creation of a new evaluation body within
the NA detailed below. Another sign of the priority given to non-legislative
activities was the stipulation that ‘during one week of sittings out of four, priority
shall be given…to the monitoring of government action and to the assessment
of public policies’ (article 48). This constituted a double shift, first, because the
government no longer held the monopoly of setting the order of the day of plenary
sessions and, second, because it promoted non-legislative debates. However, the
implementation of the provision was a source of deception. The government
(with the agreement of the parliamentary majority) largely used the possibilities of
derogation laid down in the Constitution to give priority to its own bills during
this week (Carcassonne and Guillaume, 2014). The low attendance during floor
sessions also indicated a lack of motivation among the MPs, and the ‘control
week’ was informally renamed the ‘constituency week’.
Another new article highly significant on paper was the provision of a six-week
period between the moment when a bill was first transmitted to an assembly and
its discussion on the floor. The rapporteur(s)4 and standing committee(s) therefore
had a month and a half to gather information before amending the bill. This
seemed all the more necessary given that the role of the standing committees
was comprehensively revaluated by the reform (Thiers, 2014). Indeed, the bill
considered on the floor was no longer the original one drafted by the government;
in fact, as in most European legislatures, it was the text modified by the leading
committee. As standing committees were strengthened, it seemed logical to secure
time for them. Yet, the Constitution stipulates that this new delay ‘shall not apply
if the accelerated procedure has been implemented’ (article 42). It is easy for the
government to use this accelerated procedure: only the board of the assembly
can oppose it, which is unlikely given the party discipline determined by the
majoritarian system. Indeed, this has never occurred to date. As a consequence,
both right- and left-wing governments decided to use this provision almost
systematically. The view that legislation requires knowledge and time did not
resist the pressure of day-to-day events, the new aggressive tone of the French
weekly press or the record unpopularity of Presidents Sarkozy (2007–12) and
Hollande (2012–17).
We can list other outcomes of the 2008 reform relative to the policy analysis
capacity of parliament: the increase from six to eight of the maximal number of
standing committees, the increased ease with which the opposition can set up an
inquiry committee, the possibility of requesting the legal expertise of the Council
of State on a private bill, the constitutional obligation to create European affairs
141
committees, the mission given to the Cour des comptes to ‘assist the parliament’
or the obligation for the government to attach an impact note to draft bills.
While this was far from negligible, it was not decisive regarding the capacity of
Parliament to develop a proper expertise. In addition, much like the changes listed
above, many of the new provisions have been implemented in a restrictive way.
It is easier for the opposition to propose the creation of an inquiry committee
but there are still possibilities of blocking it. Moreover, once in place, the key
position of rapporteur is given to a majority backbencher. Likewise, impact notes
are now compulsory but they are drafted by the same governmental department
that prepares the bill and no procedure has been foreseen to check their quality
(Combrade, 2014).5
Finally, one of the most important and rather overlooked aspects of the
reform probably lies in the obligation to award the chairmanship of the budget
committee to an opposition MP. This reform, made possible by the revision of
the Constitution and stipulated by the standing orders of the chambers, enables
a member of the opposition to access all official budgetary documents that he/
she wishes – and even to gain entry to official buildings. Although the budget
rapporteur, an MP from the majority, is the most important person during the
budgetary process, the chair can also benefit from the human resources within the
committee. Compared to the traditionally limited prerogatives of the opposition in
France, this new provision appears to be far from negligible (Carcassonne, 2013).
So far, it has mostly been used to stage colourful visits to the budget administration,
but it may play a role in the future, especially regarding controversies related to
the financial credibility of government commitments.
A deficit of expertise
Extra-institutional factors can also explain parliament’s inability to use and produce
policy analysis. The main factor involves how parliamentary resources are used.
The financial means of both assemblies are significant. The budget of €576,000
for the NA and €425,000 for the Senate allocated in 2015 place both assemblies
among the most well-funded in the world (NA, 2014).6 However, the capacity
of the assemblies to use these considerable means to collect independent and
reliable information is limited. The management of human expertise in particular
raises issues as seen by the examples of the clerks, assistants and MPs themselves.
In 2015, there were 1,162 clerks in the NA and 1,257 in the Senate. They
joined the assemblies after passing national examinations prepared within the best
French special institutions. Many praised their qualities as bureaucrats and their
commitment to public service. However, by recruiting young inexperienced
people who will spend their whole career within the assembly, parliament loses the
opportunity to benefit from the experience of professional seniors, and especially
that of former high-ranking civil servants from governmental departments.
Employees under contracts are rare: in early 2015, there were only 115 in the
NA. In addition, each assembly has long developed a drastic turnover system that
142
Beyond weakness
forces clerks across all positions to change services after a few years. The system
has its merits but also its flaws in terms of cumulative knowledge. It constitutes
a major source of imbalance vis-à-vis the governmental administrations where
the same people remain in similar branches for decades.
The second problem involves MPs’ assistants and their groups. Of these, 2,181
were employed in the NA in early 2015 on a full- or part-time basis, which means
3.8 contracts for each député. An MP is free to decide how to use the resources
provided for assistance. Surveys indicate that a député employs, on average, one
assistant in Paris and two to three locally (Fretel and Meimon, 2015; see also Le
Lidec, 2008; Kerrouche, 2009). Locally, employees also tend to have more full-
time contracts. In the competitive environment caused by a plurality electoral
system and multiple interchanges between the left and the right, the emphasis
is clearly put on surgery work, which is further confirmed by the vision of the
mandate shared by MPs (Costa et al, 2013). In addition, the level of expertise
in public policy issues is only one resource among others sought by MPs when
forming their staff. Party proximity, local and family roots or the network within
the party are just as, if not more, significant. As a result, assisting an MP in Paris
tends to be largely perceived as a way of entering full-time politics (Boelaert et
al, 2017). The picture is even gloomier when it comes to parliamentary party
groups. Despite their considerable budget – the NA groups globally receive
approximately €10 million a year – they employ only a handful of collaborators:
for instance, in 2016, the socialist majoritarian group of the NA had only 25
employees (excluding the communication service). Moreover, political proximity
to the party chair matters more than competence. Ultimately, it appears that party
groups tend to reproduce rather than offset the policy analysis deficit of political
parties and party foundations as suggested by this telling anecdote: in 2012, the
right-wing Union pour un mouvement populaire (UMP) group, which held the
majority at that time, secretly lent €3 million to the UMP party.7
Finally, turning to the profiles of the MPs, a few elements detrimental to the
policy analysis capacity of parliament emerge. Almost one MP out of two has
previously worked as a civil servant (François and Grossman, 2011; Rouban,
2011) – a proportion that puts the French parliament at more than ten points
above the European average. While there appears to be more civil servants on the
left than on the right, the difference between political families in this regard has
drastically decreased over the last 30 years (Costa and Kerrouche, 2007). More
recently, there has been a trend toward further professionalisation with regard to
the selection of only those MPs with a political background (Rouban, 2011; 2012;
Boelaert et al, 2017). Of the 577 députés elected in 2012, 158 (27 per cent) had
never worked outside politics. They had been assistants or advisors to ministers,
local government leaders and…MPs. Owing to the presence of both civil servants
and professional career politicians, French legislators have less experience of
some realities of contemporary France – particularly regarding the private sector
(Costa and Behm, 2013). Another important specificity of French MPs lies in
the accumulation of offices (cumul des mandats): 80 per cent to 90 per cent of
143
the members of both assemblies are simultaneously locally elected and most are
in charge of a local government. This practice makes the daily management of
their agenda more complex (Lefebvre, 2014) and contributes to their reduced
involvement in Parisian legislative or oversight parliamentary activities (Bach,
2012). Recently, a law seeking to abolish this cumul in local leadership positions
has been passed and has become effective after the general elections of 2017. This
may drastically change MPs’ behaviour in both houses.
Beyond their professional background, the gender, age and social profiles of
MPs, their assistants and the clerks also affects their capacity to perceive and
support some issues. The slow and tardy rate of change regarding gender in France
is a particularly well-studied phenomenon (Gardey, 2015; Achin and Gardey,
forthcoming). The presence of elected women remained almost the same from
1945 to 1997 – around 5 per cent of the NA. Since then, the increase has been
progressive, slow and incomplete despite the vote of affirmative action laws in
1999 and 2000. In 2012 for the first time at the NA and in 2014 in the Senate,
women represented more than one quarter of the seats, which is exactly the
European average – an improvement resulting from left-wing efforts only. The
gender unbalance is also present at the civil servants’ level since two thirds of the
200 highest clerks of the NA are male (NA, 2015) – the ratio being exactly the
same in the Senate.8 Although surveys tend to show that female MPs do not
especially focus on ‘female topics’ (Pearson and Dancey, 2012; Murray, 2010), the
deficit of women necessarily has an impact on the capacity to put some related
issues on the agenda – given the gender division of labour in many aspects of
social and economic life.
As a result of the disengagement of many MPs, the limited capacity of their
assistance and groups, and the choices made in the management and selection
of clerks, a certain amateurism can be noted in how information is processed:
there is no such thing as calls for evidence. Instead, the rapporteurs and standing
committees tend to hear actors that have already been heard at the governmental
level and by the other assembly. The representatives of the administration are
by far the most numerous. A study of 9,300 hearings mentioned in more than
a thousand NA reports published from 2007 to 2010 revealed that nearly half
of the witnesses called originated from public bodies.9 Among the civil society,
academics and experts, there is little variation regarding the participants with
the same personalities invited year after year. Over the years, some of them have
developed an impressive capacity of influence over backbenchers as indicated, for
instance, by a case study on the controversial law on the Muslim veil passed in
2010 (De Galembert, 2014). Recently, the NA has been allowed to externalise
the collection of information by contracting research centres or audit cabinets.
This decision seemingly breaks with the view that each assembly should be able
to develop its own expertise alone – a view related to the secular strategy of
autonomy of the legislative power vis-à-vis the state and even society in France
(Gardey, 2015). So far, it has only been implemented in a limited manner.
144
Beyond weakness
The discrete contribution of the French parliament to policy

analysis
Despite the many boundaries to the parliament’s ability to use and produce
knowledge relative to public policies, both assemblies often show a capacity to
develop some expertise over a variety of topics. Naturally, this concerns the issues
that directly and personally concern MPs themselves such as the organisation of
regional and local governments (Le Lidec, 2012) or the funding of political life
(François and Phélippeau, 2015). But there are also many other issues for which
the parliamentary arena is able to gather an important amount of expertise.
For instance, the last laws regarding euthanasia were passed in 2005 and 2016
through private bills after extensive consultations piloted by a couple of majority
and opposition MPs in the NA. On a very different matter, the long process of
reforming the budgetary procedure developed throughout the 2000s has largely
been initiated by specialised MPs and clerks from finance committees (Waline
et al, 2009). Across the issues, many monographs point to the appropriate, if
not high-level, expertise of MPs in various policy fields: weapons and military
operations (Rozenberg et al, 2015), foreign affairs (Riaux, 2014), civil rights
(Lascoumes, 2009) or jail policies (Chabbal, 2016). How can this be explained
given the shortcomings previously identified? To assess the actual use of policy
analysis in parliament, it is necessary to distinguish between agenda setting and
evaluation.
The French parliament and agenda setting
‘Parliamentary reports make the glory of the shelves’ is an ironical sentence

often employed in the corridors of the Palais-Bourbon. It means that although
parliament is not inactive and may produce interesting works, there is no genuine
follow-up. Many cases, however, suggest that this view is largely biased. In
November 2012, for instance, a few months after a series of terrorist attacks that
killed a number of people in Toulouse, the Green députés proposed to create an
inquiry committee on the ‘weaknesses’ of the intelligence services – a routine
kind of proposal on an unusual, if not taboo, subject. After tense debates over the
risk of threatening French defence, the inquiry committee was decided upon.
The chair of the prestigious law committee, Mr Urvoas, was elected rapporteur.
All the key actors in the field – more than 30 – were heard, often behind closed
doors. The report was finally published in May 2013 in a far less controversial
atmosphere than when the committee was launched. Two years later, in March
2015, the government proposed a bill on intelligence services that revoked most
of the proposals made in the report. Once again, Urvoas acted as rapporteur.
The bill passed consensually in both assemblies and a few months later Urvoas
became the minister for justice. This episode indicates that even in a strategic
field such as intelligence services where the executive is supposed to act more
or less discretionarily, parliament can play a decisive role. Parliament’s inquiries,
145
reports, debates and questions may matter. They can especially help set the agenda
regarding the future options taken by the government. Four specific assets are
particularly valuable in this perspective.
First, time plays an important role for parliament. Although the assemblies work
under the pressure of tightly packed agendas, they produce documents, analyses
and reports that may have an impact in the long run. It has been stated on several
occasions that a parliamentary report, seemingly forgotten just after its publication,
was taken up years later by a governmental service in charge of preparing a new
bill. An important parliamentary report published on a topic, for instance on jail
policies, is something that decision-makers cannot totally ignore – even if they
contest its conclusion. Among other elements, these types of reports may set the
agenda. The quality and depth of many reports and the prestige of the assemblies
might partly explain such influence over the long term. The continuity of some
of the views supported by specialised MPs is also important. The duration of
parliamentary careers, the principle of seniority in accessing key positions, the
cyclicality of some activities – notably the passing of the finance bill – all contribute
to give some kind of continuity to parliament. Thus, a field-expert rapporteur
working on the finance bill is in a position to raise the same concerns year after
year. From this perspective, the rotation of officials, evoked previously, is not
necessarily a weakness since it may lead a clerk freshly arrived in a new position
and in charge of a given task to refer back to the previous parliamentary report
written on the same topic. While this does not contribute to innovation, it gives
MPs’ views a coherency that can have weight in the long run.
Second, despite its weakness under the Fifth Republic, the French parliament
is still seen by policy actors as the symbolic locus of democracy. Although they
begin to defend their views long before the parliamentary stage, lobbyists,
NGOs and academics expect to be heard by committees. They never forget to
mention this in their yearly reports and boast if one of their claims is supported
by a parliamentary report. The normative status of parliament also contributes
to crediting the positions taken within it, especially when they are consensual.
From a constructivist standpoint, it can even be noted that the converging views
of left- and right-wing MPs, officially presented on a report or during a debate,
help establish the ‘win-set’ of the possible outcomes for a given issue. This is
especially true for diplomatic questions discussed in foreign affairs or European
Union committees: the seemingly pointless backbenchers’ debates contribute,
among other things, to the day-to-day elaboration of what ‘national interest’
means.
Third, parliaments are both flexible and endowed with important resources.
They offer a great variety of oversight instruments depending of the content of
the issue as well as its political sensitivity (Lazardeux, 2009). A standing committee
can independently decide to investigate a given issue, for instance by creating a
kind of inquiry sub-committee called a mission. Moreover, at the beginning of
each parliament, MPs also establish inter-party study groups on specific issues as
well as so-called friendship groups focused on parliamentary diplomatic bilateral
146
Beyond weakness
relations. Some of these groups, such as the groups on hunting activities or

private–public partnerships, have constituted, in the recent past, a key space for
lobbying. In comparison with think tanks and even governmental departments,
both assemblies also have the capacity to devote important financial and human
resources for gathering expertise. MPs are able to travel in France and abroad. The
websites of both assemblies offer access to a considerable number of documents
compared to most of the other European parliaments.
Finally, the logic of parliamentary behaviour is, at least partly, well-fitted to
gathering expertise. An MP can hope to benefit from specialising in a given topic
in the eyes of the public at large but also of parties’ gate-keepers. The example
about Urvoas cited above is telling. Being known for one’s mastery of a set of
issues may help one become a minister, not necessarily because of the expertise
accumulated, but, rather because of the notoriety attained among rank-and-file
backbenchers. Despite the fact that other resources, such as party ties, may turn
out to be more crucial for a career, existing surveys indicate that most of the MPs
share the view that ‘if you aren’t specialised, you’re dead’ (Abélès, 2001; Costa
and Kerrouche, 2007). Freshly elected MPs thus seek a niche in which they can
build a reputation for being serious and reliable, and work on their personal
network. Detailed investigations indicate that the actual level of specialisation or
commitment of a supposedly specialised MP is not necessarily as significant as for
the European affairs committee chairs for instance (Rozenberg, 2009). However,
maintaining a reputation for expertise requires some effort, especially since some
of the oversight activities are precisely those that cannot be fully delegated to
officials and assistants: hearing witnesses, travelling to the field or talking in a
committee meeting and on the floor.
Evaluating public policies from the French parliament
Legislative institutions might play a role as both a sponsor and implementer of

a given evaluation. In the US, for example, the evaluation process at the federal
level is dominated by a specific agency created in 1921 and closely related to the
Congress – the Government Accountability Office (GAO). While congressmen
may request the GAO to carry out a specific evaluation, the GAO can also
autonomously initiate the examination of the outcomes of a specific policy to
assess its normative and financial regularity. Moreover, the GAO benefits from
a real administrative capacity with an annual budget of $534 million and more
than 3,000 employees. The situation is quite different in France where there is
no real equivalent even if some recurrent initiatives have tried to institutionalise
an evaluation capacity within the French parliament (Lacouette-Fougères,
Lascoumes, 2013).
One of the first significant institutions was created in 1983 – the Office for
Scientific and Technological Assessment (Office parlementaire d’évaluation des choix
scientifiques et technologiques [OPECST]). At the time, the government justified the
creation of this organisation using the following argument: technological issues
147
had become even more complex and subject to rapid change, and it was thus
necessary to benefit from the advice of a specialised committee composed of MPs
(18 députés and 18 senators) assisted by a scientific advisory board composed of 24
experts. The OPECST, however, was also designed on the basis of the Office of
Technological Assessment (OTA) created in 1972 within the US Congress to assess
the socio-economic problems and the ethical issues related to new technologies.
Contrary to its equivalent in the US (the OTA was dismantled in 1996), the
OPECST is still functional and it publishes ten detailed reports on average per
year on key scientific issues, irrespective of whether they are related to scientific
controversies or law implementations concerning medical or scientific issues. For
instance, a recent report was devoted to assessing the so-called French model in
the regulation of ‘genetic vegetal resources’.
Although this institution is still in place, it is fair to say that its role has remained
limited. Its competencies are focused on some major bioethical issues and, more
often than not, the OPECST is mainly used to follow up some recent laws which
have had consequences for research activities. Analysing this situation, Yannick
Barthe (2002) proposed some basic explanatory factors. As in other policy domains
(Bezes, 2012), he stated that this OPECST had to compete with French high-
ranking civil servants, especially the Grands Corps, which have largely succeeded
in focusing their expertise on policy analysis. Consequently, the access to relevant
information and expertise has often been constrained and the reports published
by the OPECST have had very poor visibility and legitimacy. Second, successive
governments have always been very reluctant to grant space to controversies or
debates which might influence political mobilisation and dissent. Finally, the fact
that the OPECST is divided into two main bodies composed respectively of MPs
and experts has led to the emergence of competition between political actors
and scientific experts. According to Lacouette-Fougères and Lascoumes (2013,
59–60), many MPs fear that their deliberation will be contaminated by experts’
views, which ‘could constrain their choices and submit democratic norms and
values to rational criteria’.
The seniority of this scientific office contrasts with the great difficulties the
NA has encountered in setting up a durable structure aimed at assessing public
policies. Following the election of Jacques Chirac to the presidency in 1995,
two laws were adopted to fulfill one of the promises he had made during the
presidential campaign, that is, the modernisation of parliament. Two offices were
then created, one for evaluating public policies (Office parlementaire d’évaluation
des politiques publiques) and another for evaluating the law-making process and
legislation (Office parlementaire pour l’évaluation de la législation). Philippe Séguin,
Speaker of the NA at that time, played a leading role in this reform. However,
the division of these offices across the two chambers and a government shift in
1997 hindered any real implementation of this institutional framework. In 1997,
the new speaker of the NA, Laurent Fabius, decided to put aside these structures
which were dismantled in 2001 (Perret, 2014). Interestingly, when Philippe
Séguin became president of the Court of Auditors (Cour des Comptes) in 2004,
148
Beyond weakness
he developed a policy evaluation dynamic in this prestigious administrative body,

thereby transferring to the bureaucracy the reform principles which had failed
within parliament.
During the 2000s, several other specialised structures were created to assess
public policies related to specific policy fields such as the national budget with a
structure for assessment and control (Mission d’évaluation et de contrôle [MEC]),10
the welfare budget (Mission d’évaluation et de contrôle de la sécurité sociale)11 or
health policies (Office parlementaire d’évaluation des politiques de santé).12 The major
shift of the national budget procedure that started with a 2001 law (known by
the French acronym LOLF) triggered these changes. Foreign models were also
explicit references for institutional entrepreneurs, such as the National Audit
Office in Westminster in the case of the MEC. The innovative feature of these
developments contrasts with their lack of public visibility. Although some of
them continue to produce in-depth reports and, especially, to conduct numerous
hearings, they suffer from the lack of commitment of high-profile or ambitious
MPs. In a sense, both assemblies have developed a parliamentary bureaucracy
aimed at assessing the governmental one.A more advanced and major step toward
the institutionalisation of policy evaluation was taken in 2009 with the creation
of a Commission for the Assessment and Monitoring of Public Policies (Comité
d’évaluation et de contrôle des politiques publiques [CEC])13 within the NA. Although
the decision to create this body originated in the constitutional revision of 2008,
there had been tense internal disputes within the right-majority about who
would lead the reform. These tensions reveal how salient organisational issues are
within assemblies – it appears that protecting a standing committee status might
take the place of actually influencing public policies. They also confirm that the
degree of heterogeneity of a majority constitutes a major factor for institutional
innovation (Lazardeux, 2009). It was ultimately decided that this commission
would be chaired by the NA Speaker and would be partially formed of the
standing committee chairs as well as the parliamentary party group presidents.
The involvement of the standing committee leaders aimed at reassuring them
that the new structure would not deprive their committee of their prerogatives.
The high standing of the members was also supposed to demonstrate the depth
of the mobilisation of the entire chamber. However, the relative marginality of
this structure seven years after its creation indicates that it may have produced
the opposite effect. Indeed, it appears that the ‘VIP MPs’ of the commission
have other priorities beyond assessing public policies and that most of the work
is carried out by the 15 remaining backbencher members.
The balance sheet of this new structure is not yet negative. Approximately four
in-depth reports are discussed and published by the commission each year.14 They
are always co-signed by MPs from the majority and the opposition in order to
depoliticise public policy assessment. In practice, the cooperation between the
two camps is made possible by the delegation of the greater part of the writing
to clerks and by the relative confidentiality of the work. The reports deal with
a wide variety of issues. The initial official view that they were supposed to
149
consider transversal questions in order to avoid encroaching on the prerogatives

of the standing committees has not diverted the commission from considering
significant specific issues such as the training policy or anti-tobacco health policies.
In addition, other reports concerning the implementation of the recommendations
of previous CEC reports are published regularly, including some from previous
terms. Such continuity is noticeable in an institution like the NA which must
frequently respond to emergencies and topical events.
To finish, it should be noted that the parliament is also able to develop a capacity
of evaluation in some policy areas in which MPs expressed a specific interest and
expertise. This is for instance the case regarding the issue of women representation
in politics for which both assemblies have shown remarkable and continuous
interest despite changing majorities (Achin and Bereni, 2013). In 1999, a private
bill proposed by the Speaker of the NA allowed each chamber to create ad-hoc
non-legislative committees on women rights (Délégations aux droits des femmes).
Those committees – to which very few men belong – have been active since
then, without interruption, and have held an average of 20 meetings a year in the
NA. Successive reports on the implementation of the 2000 law in favour of equal
representation of male and female in politics have been crucial in understanding
the limit to this affirmative action policy. As such, they have contributed to the
regular increase of financial sanctions against macho political parties. In this case,
the capacity to evaluate the implementation of policy reforms depends crucially
on the following two dimensions: the capacity of some female interested MPs
to agree on shared diagnostic beyond party lines, and the capacity of those
committees to liaise both with NGOs and a rather weak ministerial department.
Conclusion
In conclusion, the assessment of the policy analysis capacity of the French
parliament is inevitably balanced. Beyond the well-known constitutional
constraints, the parliament suffers from a lack of capacity, willingness and know-
how in terms of collecting original, new and diverse expertise on public policies.
The fragmentation of the parliament, the influence of technocratic elites and the
relative lack of interest of the MPs largely explain why most evaluation in France
is still undertaken by the government bureaucracy as opposed to the political
control of the Congress in the US which is specifically based on its capacity for
policy analysis with the help of independent agencies such as the GAO.
Noticeable efforts have been made to offset the situation; for instance,
constitutional changes were made in 2008 and a new body for evaluation was
created in 2009 in the NA. Not everything about them is negative; indeed, some
changes introduced on these occasions may produce results in the long run. For
instance, the introduction of the impact assessment obligations in 2009 is probably
a first step toward the strengthening of a more constraining system in the wake
of the better regulation agenda supported by international organisations.
150
Beyond weakness
To date, however, these recent changes have not dramatically altered the policy
analysis deficit of the French parliament. This might largely be explained by
the lack of political incentives for MPs to specialise in policy fields, participate
in evidence collection and be involved in policy evaluation. Opposition MPs
are in an especially difficult situation given the depoliticised features of many
of these activities, such as those of the CEC. The abstract legal view that the
parliamentary evaluation of the administration should have nothing to do with
the political oversight of the government is a noticeable demotivating factor
(Rozenberg, 2013). A major stake for the future thus lies in the capacity of the
French parliament to grasp policy analysis while continuing to offer what is
expected of Parliament: pluralist debate.
Notes
1 www2.assemblee-nationale.fr/14/statistiques-de-l-activite-parlementaire.
2 Decision 59-2, 24 June 1959.
3 Decision 2009-581, 25 June 2009, point 58.
4 The rapporteur is the MP in charge of presenting a report on a given bill within a standing
committee and then of supporting the committee’s views at the floor stage. At the NA, he or
she always belongs to the majority.
5 See also his PhD thesis defended at the University Paris I in 2015.
6 Same source for the other figures of this part; Senate: Rapport d’information du Sénat 618, 20
May 2016. For a worldwide comparison: IPU, 2012.
7 ‘UMP: l’emprunt caché de 3 millions d’euros’, Mediapart, 21 June 2014, https://www.mediapart.
fr/journal/france/210614/ump-lemprunt-cache-de-3-millions-deuros?onglet=full.
8 Source: Senate documents, 31 December 2016.
9 See www.regardscitoyens.org/transparence-france/etude-lobbying/.
10 Created in 1999 at the NA.
11 Created in 2007 at the NA.
12 Suppressed in 2009 when the CEC was created.
13 See the ongoing PhD thesis prepared at the University Paris II by Hortense de Padirac. We
thank her for her valuable comments.
14 In addition, the CEC can also assess the assessment notes prepared by the government and
accompanying bills. So far, this activity has not been developed.
References
Abélès, M. (2001) Un ethnologue à l’Assemblée, Paris: O. Jacob.
Achin, C., Bereni, L. (2013) ‘Comment le genre vint à la science politique’, in
Dictionnaire genre and science politique: Concepts, objets, problèmes, pp. 13–41, Paris:
Presses de Science Po.
Achin, C., Gardey, D. (forthcoming) ‘Genre et Parlement’, in O. Rozenberg, E.
Thiers (eds) Traité d’études parlementaires, Brussels: Larcier.
Bach, L. (2012) ‘Faut-il abolir le cumul des mandats?’, Paris: Editions Rue d’Ulm.
Barthe, Y. (2002), Le rôle de l’OPECST, une mise en perspective historique, Rapport
pour le programme ‘Concertation, décision, environnement’, Grenoble:
CERAT-CSI.
151
Bezes, P. (2012) ‘État, experts et savoirs néo-managériaux: Les producteurs et

Boelaert, J., Ollion, E., Michon, S. (2017) Métier: député, Paris: Raisons d’agir.
Carcassonne, G. (2013) ‘L’opposition parlementaire comme objet juridique:
Une reconnaissance progressive’, in O. Rozenberg, E. Thiers (eds) L’opposition
parlementaire, pp. 85–94, Paris: La Documentation française.
Carcassonne, G., Guillaume, M. (2014) La Constitution (12th edn), Paris: Seuil.
Chabbal, J. (2016) Changer la prison, Rennes: Presses universitaires de Rennes.
Combrade, B.L. (2014) ‘L’étude d’impact au Parlement français: Un instrument
de mutation du rôle des assemblées dans le processus législatif ?’, Revue française
d’administration publique 149(1), 195–206.
Costa, O., Behm, A.-S. (2013) ‘Les députés connaissent-ils l’entreprise?’, En temps
réel: Les cahiers, 48. Available at: http://www.entempsreel.com/files/cahier52.pdf
Costa, O., Kerrouche, E. (2007) Qui sont les députés français: Enquête sur des élites
inconnues, Paris: Presses de Sciences Po.
Costa, O., Brouard, S., Kerrouche, E., Schnatterer, T. (2013) ‘Why do French
MPs focus more on constituency work than on parliamentary work?’, Journal
of Legislative Studies 19(2), 141–59.
De Galembert, C. (2014) ‘Forcer le droit à parler contre la burqa: Une judicial
politics à la française?’, Revue française de science politique 64(4), 647–68.
Döring, H. (ed) (1995) Parliaments and Majority Rule in Western Europe, Frankfurt:
Campus Verlag.
Elgie, R. (2011) Semi-Presidentialism: Sub-Types and Democratic Performance, Oxford:
Oxford Universiy Press.
François, A., Grossman, E. (2011) ‘Who are the deputies of the Fifth Republic?
Some figures’, French Politics 9(4), 364–80.
François, A., Phélippeau, E. (2015) Le financement de la vie politique. Réglementations,
pratiques et effets politiques, Paris: Armand Collin.
François, B. (2007) ‘Le Parlement sous la Ve République’, in J. Garrigues (ed)
Histoire du Parlement de 1789 à nos jours, pp. 427–93, Paris: Armand Colin–
Assemblée nationale.
Fretel, J., Meimon, J. (2015) ‘Les collaborateurs parlementaires, un vivier pour
les cabinets ministériels?’, in J.-M. Eymeri-Douzans, X. Bioy, S. Mouton (eds)
Le règne des entourages: Cabinets et conseillers de l’exécutif, pp. 527–51, Paris: Presse
de Sciences Po.
Gardey, D. (2015) Le linge du Palais-Bourbon: Corps, matérialité et genre du politique,
Lormont: Le Bord de l’Eau.
Gicquel, J., Gicquel, J.-E. (2015) Droit constitutionnel et institutions politiques, Paris:
Librairie Générale de Droit et de Jurisprudence (LGDJ).
Hayward, J. (2004) ‘Parliament and the French government’s domination of the
legislative process’, The Journal of Legislative Studies 10(2), 79–97.
Huber, J.D. (1996) Rationalizing Parliament, Legislative Institutions and Party Politics
in France, Cambridge: Cambridge University Press.
152
Beyond weakness
IPU (Inter-parlementary Union) (2012) ‘The changing nature of representation’,

Global Parliamentary Report, Geneva: IPU.
Kerrouche, E. (2009) ‘Usages et usagers de la permanence du député’, Revue
Krehbiel, K. (1991) Information and Legislative Organization, Ann Arbor, MI: The
University of Michigan Press.
Lacouette-Fougères, C., Lascoumes, P. (2013) ‘“Les scènes multiples de
l’évaluation”: Les problèmes récurrents de son institutionnalisation’, Laboratoire
interdisciplinaire d’évaluation des politiques publiques (LIEPP) Policy Paper 1, Paris:
LIEPP-Sciences Po.
Lascoumes, P. (2009) ‘Les compromis parlementaires, combinaisons de
surpolitisation et de sous-politisation’, Revue française de science politique
59(3), 455–78.
Lazardeux, S. (2009) ‘The French National Assembly’s oversight of the executive:
Changing role, partisanship and intra-majority conflict’, West European Politics
32(2), 287–309.
Le Lidec, P. (2008) ‘Les députés, leurs assistants et les usages du crédit collaborateurs:
Une sociologie du travail politique’, Sociologie du travail 50, 147–68.
Le Lidec, P. (2012) ‘Decentralisation and territorial reforms in France: How
constitutional constraints impact strategies for reform’, in A. Benz, F. Knüpling
(eds) Changing Federal Constitutions: Lessons from International Comparison, pp.
249–67, Opladen: Verlag Barbara Budrich.
Lefebvre, R. (2014) ‘Les élus comme entrepreneurs de temps: Les agendas des
cumulants’, in D. Demazière, P. Le Lidec (eds) Les mondes du travail politique:
Les élus et leurs entourages, pp. 53–70, Rennes: Presses universitaires de Rennes.
Murray, R. (2010) ‘Second among unequals? A study of whether France’s quota
women are up to the job’, Politics and Gender 6, 93–118.
NA (National Assembly) (2014) Annexe au rapport de certification de la Cours des
Comptes, Paris: NA
NA (National Assembly) (2015) Bilan social, Paris: NA.
Norton, P. (1990) ‘Conclusion: Legislatures in perspective’, in P. Norton (ed)
Parliaments in Western Europe, pp. 143–52, London: Frank Cass.
Norton, P. (ed) (2002) Parliaments and Citizens in Western Europe, London:
Routledge.
Pearson, K., Dancey, L. (2012) ‘Speaking for the under-represented in the House
of Representatives: Voicing women’s interests in a partisan era’, Politics and
Gender 7, 493–519.
Perret, B. (2014) L’évaluation des politiques publiques, Paris: La Découverte.
Pouvoirs (2013) ‘Le renouveau du Parlement’, 146.
Riaux, G. (2014) ‘Les professionnels de la politique et la politique étrangère’,
Gouvernement et action publique 3(1), 51–73.
Rosanvallon, P. (2015) Le bon gouvernement, Paris: Seuil.
153
Rouban, L. (2011) ‘Les députés’, Les Cahiers du CEVIPOF n°55, Paris: Sciences
Po, Centre de recherches politiques. Available at: http://www.cevipof.com/
fichier/p_publication/829/publication_pdf_cahier_55.3_jp.pdf
Rouban, L. (2012) ‘Les députés de 2012: Quelle diversité?’, Elections 2012:
Les électorats politiques 8, Note de recherche du CEVIPOF, Paris: Sciences Po,
Centre de recherches politiques. Available at: http://www.cevipof.com/fr/
les-publications/les-cahiers-du-cevipof/bdd/publication/829
Rozenberg, O. (2009) ‘Présider par plaisir: L’examen des affaires européennes
à l’Assemblée nationale et à la Chambre des Communes depuis Maastricht’,
Revue française de science politique 59(3), 401–27.
Rozenberg, O. (2013) ‘L’opposition parlementaire, espèce à protéger’, in O.
Rozenberg, E. Thiers (eds) L’opposition parlementaire, pp. 191–210, Paris: La
Documentation française.
Rozenberg, O. (2016) ‘Un petit pas pour le Parlement, un grand pour la Vème
République’, The Laboratory for Interdisciplinary Evaluation of Public Policies (LIEPP)
Working Paper 61, Paris: LIEPP-Sciences Po.
Rozenberg, O., Chopin, O., Hoeffler, C., Irondelle, B., Joana, J. (2015) ‘Des
députés experts militaires? Les motivations et rétributions au sien des commissions
défense de parlements européens’, Politique européenne 48, 178–200.
Sieberer, U. (2011) ‘The institutional power of western European parliaments:
A multidimensional analysis’, West European Politics 34(4), 731–54.
Thiers, E. (2014) ‘Les commissions permanentes de l’Assemblée nationale et
l’élaboration de la loi depuis 2008: Une révolution très discrète’, Revue juridique
Thémis de l’Université de Montréal (RJTUM) 48(1), 211–22.
Waline, C., Desrousseaux, P., Pellé, B. (2009) Contrôle et évaluation des finances
publiques, Paris: La Documentation française.
154
Part Three
Committees, public inquiries,
and consultants
155
NINE
Public inquiries, committees

Cécile Blatrix and Guillaume Gourgues
Over the last 30 years, the proliferation of public participation mechanisms in

decision-making has led to the emergence of extremely varied participation
forms: public inquiries, users’ committees, public hearings, consensus conferences,
participatory budgeting for example. Few studies, however, have questioned
what, if anything, these mechanisms have changed in the existing dynamics of
knowledge production about and for public policy. Naturally, studies on public
participation have revolved around the capacity of mechanisms to influence the
decision-making process. Indeed, while it is far more significant to focus on how
public participation mechanisms influence the production of knowledge and
its mobilisation for public policy, this issue remains difficult to grasp. Analysing
the extent to which the knowledge produced in these mechanisms sustains the
decision-making process remains a minor concern. Put differently, the manner
in which this knowledge is taken into account in policy-making has attracted
little interest. On the contrary, attention has focused on what the existence and
development of these mechanisms change in terms of mobilising knowledge on
policy. Moreover, there has been considerable interest in the extent to which
these mechanisms are perceived as new resources or new constraints, and for
whom, even while the boundary between these two types of analyses may not
always be obvious.
Over the years, academic literature has explored this approach to public policy
(Fischer, 2000; Hisschemöller et al, 2001). Drawing on existent literature, this
chapter will focus on public inquiries and committees. We must first, however,
clarify what we refer to as ‘public inquiries’ and ‘committees’. In the French
context these two terms have specific meanings and are associated with two
distinct public participation mechanisms: while the former are geared to the
general public and mainly involve environment-related projects, the latter are
geared to those who affect or may be affected by the policies, and may be
organised across sectors and at different levels; they are commonly referred to as
‘consultative government’ or ‘democracy in public administration’ (Weber, 1968;
Le Clainche, 2011). Indeed, while the rhetoric around these mechanisms suggests
that they seek to involve ‘citizens’, ‘beneficiaries’ or the ‘target population’ in
policy-making, public inquiries and committees actually bring together different
actors whose roles have been analysed in this book: interest groups,1 NGOs2 and,
more sporadically, ‘unorganised’ publics.
We thus believe that public participation mechanisms are defined by their
public nature. ‘Public’ here refers to two specific things: first, these mechanisms
157
are involved in ‘creating public awareness’ in relation to the decision-making

process and the knowledge on which it is based; they thus make the process more
transparent. Second, the mechanisms all directly involve the target population in
knowledge production for public policy, that is, traditional representatives do not
act as mediators. Their development thus seems to reflect their desire to widen the
decision-making circle across several policy areas (Barnes et al, 2007), as has been
extensively discussed in the literature (Fischer, 2012; Fung and Wright, 2003).
In this chapter, we argue that public participation in policy analysis must
be put into perspective. We assert that, in France, the so-called ‘participatory
mechanisms’ such as public surveys and users’ committees rarely produce their
own policy knowledge. In other words, these mechanisms do not create otherwise
non-existent knowledge. While some mechanisms produce ‘original’ knowledge,
assimilating this knowledge into the policy process is far from easy. Focusing
on the French case and its specificity, we suggest that participatory mechanisms
enable both policy actors and the government to indiscriminately use knowledge,
opinions and actors from outside the state apparatus. These mechanisms, however,
may also give rise to assessments seeking to identify how public issues are
addressed. These assessments may modernise the dominant frameworks and, in
rare cases, lead to the emergence of new ‘expertise’. Participatory mechanisms
may thus enable barely visible knowledge and analytical frameworks to come to
the fore. They are occasionally involved in the interpretation, dissemination and
comparison of the knowledge produced by highly varying actors ranging from
professional interest groups to local residents. Beyond the influence on policy
analysis, one of the most significant effects in relation to the issues we raise in
this chapter is that participatory mechanisms undeniably produce more or less
formalised knowledge of the policy process itself.
Knowledge mobilisation mechanisms beyond the political and

administrative system
It seems necessary to examine why participatory mechanisms tend to develop
unevenly across policy areas. Not all policy areas are equally affected by the
tendency to include organised or unorganised civil society in these types of
mechanisms. This suggests that some sectors are more willing than others to
mobilise participatory mechanisms and the knowledge which these may provide.
For instance, while environmental policies are characterised by the emergence of
public debate forums, macroeconomic policies have remained highly monopolised
by small circles of ‘experts’, to the extent that citizens’ influence on these policies
now seems to have disappeared (Streeck and Schafer, 2013). It therefore seems
appropriate to pay more attention to the specific configurations in which these
mechanisms are created and used. Drawing on two case studies in France, this
chapter will focus on the policies in two specific sectors: the environment
and rail transport sectors. These case studies suggest that most participatory
mechanisms are characterised by a similar process: the manner in which they
158
are designed, implemented and function in these two sectors is closely linked to
the changes in power relations both within and beyond the circle of influence
of the institutional actors traditionally responsible for policy-making. A similar
process can be observed in other countries; in Britain for instance, the state’s new
system of governance known as ‘new public management’ has gradually replaced
‘public participation’ (Pratchett, 1999; Lowndes et al, 2001). However, this
process is particularly marked in France where state institutions have historically
monopolised expertise.
Indeed, since the 1980s, France has implemented a variety of mechanisms
largely analysed as participatory democracy or deliberative democracy. While
these mechanisms encourage participation in the debates and dialogue preceding
decision-making, few actors are involved in the actual decision-making process
itself. The studies that have focused on the analysis of these mechanisms in
France have become increasingly specialised, as has been observed in other
national contexts (Chilvers, 2008); little attention has thus been paid to how
public participation mechanisms may be integrated within public policy. These
mechanisms are often analysed as issues of interest in themselves, rather than as
spaces where policy analysis occurs, despite the fact that their very existence
provides valuable insights into public policy.
The growth of these different forms of participatory mechanisms suggests that
the French government, far from its traditional image as an ultra-centralised
policymaker, has become conscious of the need to strengthen its decisions by
making use of all the available knowledge. The question that arises, then, is
whether, from the perspective of national and local decision-makers, public
participation mechanisms can produce ‘new’ and ‘specific’ knowledge about
and for public policy. Caution, however, must be exercised when addressing
this question: indeed, participatory mechanisms go beyond complementing
government knowledge as they are the result of the random encounter between
different factors such as the quest for acceptability, political communication,
the attractiveness of democratic innovations and legal obligations. Analysing the
origins of public participation mechanisms reveals that knowledge production
about/for public policy was never the initial intention; rather, these mechanisms
had multiple objectives that sought to avoid conflict, reinforce decision-makers’
legitimacy and inform decision-making. Participatory mechanisms thus fall within
a context of fragmented knowledge production about policy. They are exploited
by government authorities eager to influence how this expertise circulates in an
attempt to reinforce their own position and potential monopoly.
The case of environmental policy
The use of participation mechanisms in France has been most apparent in the
fields of urban policy, transportation planning and environmental and quality-
of-life policies; it is in these fields as well that public participation has been
institutionalised. Participatory mechanisms in the environmental and urban
159
planning and development sectors are thus recognised as equal participants in

the decision-making process. Globally, the need for public participation in the
early stages of the decision-making process has been increasingly recognised in
the environmental field since the 1992 Rio Declaration and the 1998 Aarhus
Convention. Adopted after the signing of the Aarhus Convention, the 2005
French Environmental Charter, and Article 7 in particular, gave everyone the right
to participate in all national and local public decisions affecting the environment.
Against this background, and after successive reforms,3 public inquiry thus begun
to gain in importance. Public inquiry is undoubtedly the most long-standing and
most widespread form of public participation in decision-making. However, it
remains relatively unknown. Introduced in the nineteenth century to protect
private property before expropriation, public inquiry developed and was extended
to other sectors following two major reforms in 1976 and 1983.
In 1976, that is, five years after the creation of the French Ministry of the
Environment, André Fosset Minister of the Quality of Life, expressed the will to
recognise the role of associations in policy-making. As a result, for the first time,
environmental organisations were invited to participate in various consultative
committees such as the agency for waste recycling and recovery, site committees
and the executive boards of national parks and water supply agencies.
The simultaneous recognition of the environment as a service of general interest,
and of citizen participation as a guarantor of this concern, was legally established
by the law on the protection of nature on 10 July 1976. This law assigned a new
task to public inquiry4 requiring it to make environmental impact assessments
available to the public; this became obligatory when projects touched on the
environment. Publishing impact assessments was perceived as a means of ensuring
their quality and reliability by forcing their developers to anticipate controversies.
In 1983, the scope of public inquiries was extended even further: public
inquiry was now expected to consider private property, strengthen environmental
protection and democratise decision-making, as evidenced by the title of
the Bouchardeau law of 12 July 1983 on the ‘democratisation of public
inquiries and environmental protection’. Since the reform, 15,000 to 20,000
projects have been subjected to public inquiry each year.5 Public inquiries are
generally organised as follows. Supervised by a head of inquiry (commissaire
enquêteur ) or by a commission of inquiry (commission d’enquête), inquiries usually
last for one month. During this period, the public is free to view documents
relating to the project (including environmental impact assessments) and record
their observations in a register available in the town halls of each municipality
in which the project is to be undertaken. Once complete, the head of inquiry
drafts a report in which he/she draws conclusions which can be favourable,
unfavourable or favourable but with reservations. In contrast to public inquiries
in Britain, French public inquiries are thus essentially a written and very formal
procedure. Unlike their British counterparts, French heads of inquiry are not
trained professionals who are regularly assessed but, rather, occasional collaborators
seconded from government.
160
Despite successive reforms, much criticism continued to be levelled against

public inquiry. Indeed, this process occurred rather late in the decision-making
process and was thus referred to, notably by environmental organisations and
activists, as a ‘mock democracy’ set up to endorse decisions already taken at
a higher level. This criticism paved the way for ‘public debate’ in the 1990s;
organised earlier in the process, public debate was implemented for the largest
projects. Such debates provided a forum making it possible to discuss the very
relevance of projects. Subsequently, as debate on the decision-making process
progressed, this issue was considered to be fully resolved. Making environmental
impact assessments public via public inquiry enabled associations to act as potential
whistle-blowers. For the young Ministry of Environment established in 1971, this
development represented a great opportunity as it made it possible to strengthen
the role played by the public in general and, specifically, the role of environmental
organisations. This was perceived as the best way to protect the environment and
mobilise various forms of support from appropriate arenas when necessary. It is a
well-known fact that environmental organisations had long acted as ‘decentralised
services’ for a Ministry that lacked these services over a long period of time
(Lascoumes, 1994). From the emergence of the environmental policy, those
responsible for the structurally weak Ministry of Environment thus endeavoured
to maintain a close relationship with these ‘supporters’ (Spanou, 1991). Within the
context of neo-corporatism à la française (Muller, 1984) characterised by markedly
strong sectoral dynamics, the Ministry of the Environment was particularly
deprived of resources. Indeed, expertise had been independently produced by
the administrative bodies of other ministries and, as such, environmental issues
lacked dedicated policy analysis. Consequently, since its establishment in 1971,
the Ministry of the Environment has drawn its knowledge from pre-existent
major technical institutions: the institute of Bridges (corps des Ponts), the Water
and Forestry institute attached to the Ministry of Equipment and Transport,
and the Ministry of Agriculture and Forestry and Mines for projects on soil and
energy. Unlike other sectors, the environmental policy does not therefore rely on
its own elite institutions, that is, on the specialised state bodies enjoyed by other
ministries. While this option had been initially considered, it was finally rejected
(Kessler, 1999). Many Environment ministers have described the difficulties
which they encounter when dealing with sectors which have specialised bodies at
their disposal (see Lepage (1998) for a vivid account of one former Environment
Minister). In an attempt to find its place among other more established policy
sectors, the environmental policy began to rely on environmental groups, thus
giving them a platform to express themselves. The development of an ‘advisory
board’ and participatory tools thus gave a voice, and a role, to these more or
less institutionalised environmental organisations which provided the ‘support’
mentioned earlier. These organisations’ demands for greater public participation
have always been tied to the defence of their inclusion in the development of
projects and decision-making.
161
Regional rail transport policies
The establishment of users’ committees, known as railway committees (comités de

ligne), in the early 2000s to support new regional railway policies also made the
same strategic use of associative knowledge. After a five-year period during which
experiments were conducted in seven regions, the management of the passenger
railway service was transferred to French regional councils in 2002. To determine
their policy content, that is, the services offered, investment priorities and pricing
strategy, these councils were expected to negotiate with the traditional national
railway operator (SNCF), a monopoly that enjoyed a very strong position in rail
expertise. However, each regional council had already set up a users’ committee
between 1999 and 2006. Initiated in Alsace in 1997 and legalised in 2000, these
comités de ligne could be described as consultative bodies set up by regional councils
on a voluntary basis.6 They brought together various stakeholders such as local
elected officials, users (individual or community representatives) and railway
unions to discuss the development of the regional transport offer in single railway
lines or specific sections of the network. While they differed in name, frequency
and in how they monitored proposals, all these users’ committees were actually
one and the same mechanism, that is, comités de ligne. The popularity of users’
committees among regional councils today reflects the power struggles specific
to the railway sector where regions must assert their legitimacy and credibility.
An analysis of how these comités de ligne operate (Gourgues, 2012) reveals how
they produce knowledge. Indeed, users’ associations use these mechanisms the
most. Specifically, irrespective of the region analysed, there appears to be a small
core group of association leaders who regularly attend these committees. This
is despite the fact that these associations have considerable expertise in transport
issues because of their members’ trajectories (former railway workers, SNCF
management, or trainspotters commuting by rail over the years), the existence of
nationwide federations, and their ability to give voice to users’ recurring requests
(punctuality, schedules, information displays at stations, and so on). This expertise
allows them to publicly initiate discussions with the SNCF, which can be technical
and even highly specific. Discussions revolve around the problems encountered
with specific railway lines such as electrification, equipment available and possible
technical solutions. These associations possess alternative knowledge which
occasionally enables them to challenge the SNCF’s expertise and demand answers
and commitments; specialised agents from regional councils can in turn draw on
this knowledge. Several consequences have ensued from the use of ‘comités de ligne’.
First, these associations have marginalised the participation of non-members who
generally lack the technical skills required to participate in discussions, even when
these discussions focus on their problems and their demands. Users seeking to be
active participants gradually join other associations which allow them to acquire
a sufficient level of expertise. Second, the slowdown of regional investments has
been put on the agenda. Indeed, while regional councils sought to increase their
legitimacy by relying on expertise from beyond the sphere of the SNCF during
162
periods of sustained investment (2002–10), the rapid decline of regional interest

in these committees has been associated with the sharp slowdown of investment
that began in 2010. The participants who had actively participated in discussions
thus found it difficult to deal with the progressive lack of interest and challenged
the decreasing budget.
In both environmental and regional railway policies, public participation
mechanisms thus operate as spaces where knowledge and actors are mobilised.
They have hence been able to establish a fragile form of legitimacy. Thus has
emerged the crucial role of uncertainty that Michel Callon, Pierre Lascoumes
and Yannick Barthe (2013) observed when alternative knowledge was sought
by challenging expertise. It seems important, however, to underscore that real
alternative knowledge and the de-monopolisation of expertise remain closely
controlled and relatively constrained within participatory mechanisms. It is not
so much a matter of producing knowledge than of mobilising and bringing to
the fore cognitive and human resources in terms of support and the ability to
mobilise people in the context of emerging policies whose legitimacy is yet to be
asserted. Irrespective of whether they are public inquiries or users’ committees,
French participatory mechanisms do not produce knowledge in the strictest
sense. To understand how they participate in policy analysis, these mechanisms
must be analysed as part of a system of interacting institutions, each fighting to
impose its own vision of decision-making.
While the academic literature has argued that involving the public in policy-
making may produce new knowledge to inform decision-making, we found
that this view was challenged by those responsible for knowledge production.
Moreover, strong advocates of participatory mechanisms were often hard put to
give specific examples of these ‘nuggets’ of participation. Indeed, admitting that
lay citizens might produce useful knowledge hitherto unknown to the institutions
responsible for knowledge production and to policy officials is tantamount to
admitting that the expertise is faulty and that the representatives barely represent
the concerned parties. This clearly shows that public participation mechanisms
challenge the very production and mobilisation of knowledge, as well as its
reliability, solidity and even its ‘righteousness’.
Public participation thus relies on a fine and hesitant balancing act between
giving people greater access to the decision-making process on the one hand,
and controlling this access on the other. Why, then, do public officials engage
in setting up mechanisms they paradoxically define as useless or dangerous?
The answer lies in the power struggles taking place among these officials. Civil
servants, for instance, attempt to free themselves from policy-making through
public participation and the use of dynamics very similar to what is observed
when a new administration is put in charge of producing studies (Mazoyer,
2012): when attempting to limit the influence of administrative departments
thought to wield excessive power, policy officials turn to public consultations and
committees to obtain valuable support, especially as these can often unilaterally
define organisational design. Conversely, mayors can also rely on users’ or residents’
163
committees to urge the different departments to change policies. For instance,

a Minister might organise a widespread consultation with farmers in the form
of a General Assembly for Agricultural Development to stand up to primarily
hostile agricultural associations (Suaud, 1984).
Local authorities can rely on users’ committees to impose guidelines in relation
to new policies when faced with operators who hold most of the expertise. With
regard to knowledge production, competition is therefore rife and occurs within
the context of consultations and committees.
These mechanisms are used and are sometimes created amid institutional
struggles between the different administrative and policy sectors. However,
while strategic uses by ‘official’ actors exist, little is done to prevent the actors
from overstepping the boundaries imposed, especially if they concern social
mobilisation.
The increasing complexity of the decision-making process and

competing forms of expertise
A few French policy studies assert that understanding how target populations
perceive government knowledge is essential. They argue that, ‘depending on the
contexts and the publics involved, knowledge mobilisation can either contribute to
legitimising government bureaucracy or trigger criticism’ (Bongrand et al, 2012).
Similarly, studies on the coalition and protest logics at work in environmental
assessment worldwide have revealed that the influence produced by knowledge
within participatory mechanisms in relation to public decisions depends largely on
the power relations established through social mobilisation (Devlin, Yap, 2008).
Indeed, how these mechanisms influence knowledge production and mobilisation
in policy has remained somewhat ambivalent. In this second section, we will
distinguish the different types of potential knowledge mobilisation within the
framework of participation mechanisms, irrespective of their origins; indeed,
these origins may strengthen decision-makers’ legitimacy or, on the contrary,
challenge or outflank them.
The mobilisation of unrepresented lay knowledge in the knowledge production

system
Many participatory mechanisms have failed to produce entirely new knowledge

on public policy. However, some authors, especially in France, argue that lay
citizens may generate ‘lay knowledge’ which can be distinguished from traditional
forms of expertise. The development of public consultation mechanisms is part
of a trend which questions the restricted development of scientific and technical
expertise (Wynne, 1996; Lascoumes, 2002; Akrich et al, 2010; Barbier, 2013); the
existence of ‘dispersed knowledge’ (Sunstein, 2006) thus plays a key role insofar
as it justifies the participation of the ‘general’ or ‘lay’ public.
164
Michel Callon (1998) has distinguished three models to characterise the

relationship between science and society. The first is the ‘public instruction’ model
which supports the view that ‘not only must scientists teach the public everything,
they can learn nothing from the public’. The second is the ‘public debate’ model
in which scientific knowledge is complemented by local, increasingly complex
and ever-changing knowledge. Callon argues that public inquiries and public
hearings fall within this second model as they rely on comparisons between
different viewpoints, forms of knowledge and judgements which, despite their
differences, mutually enrich one another. The third is the ‘co-production of
knowledge’ model in which scientific expertise not only asks for but also integrates
the public’s opinion ‘as soon as the actors start developing knowledge affecting
the concerned public’.
The co-production of knowledge has received particular attention in sciences
studies (Fiorino, 1990) and has also attracted the interest of French urban studies.
For instance, Héloïse Nez and Yves Sintomer (2013) have proposed a global
typology of the type of knowledge lay citizens participating in policy analysis
can produce. These authors have distinguished three types of knowledge: ‘local
knowledge’ referring to individuals’ ‘common sense’ or concrete collective
experiences; ‘expert or professional knowledge’ linked to an individual’s specific
skills or to the production of a second, independent, expert opinion; and ‘political
knowledge’ based on individual and collective activism. Naturally, this typology of
produced knowledge is politically significant insofar as it is expected to assess the
extent to which citizens might contribute to policy-making. Similar expectations
can be found at the international level as evidenced by the approaches that focus
on evidence-based policies (Corburn, 2005; Cartwright, 2007).
There are other types of knowledge related to the contexts of victimised actors
or those excluded from the decision-making process. Participatory mechanisms
may thus occasionally allow actors relatively excluded from the political process
to express themselves: for instance, the priorities and perceptions of a category
of the population (the inhabitants of a neighbourhood, youth, elderly or disabled
people) can be taken into account to select elements that may be of interest to
public policy. A wide range of committees such as youth committees, elders’
committees and neighbourhood committees are in line with this approach. Other
mechanisms have sought to include the actors affected by the decisions taken.
Public inquiry in France has largely been conceived as a means of preventing
future conflicts. Specifically, it has been designed to avoid conflicts over private
property issues when projects presumed to be in the general interest are being
developed. The ‘taking into account of collateral damage’ thus complements
more traditional expropriation processes.
These participatory mechanisms thus seek to acknowledge those whom new
infrastructure or projects will not directly benefit but who will nonetheless
have to bear the inconveniences. On the one hand are thus the beneficiaries of
public policy and on the other hand its victims, who are expected to sacrifice
their private interests in the interest of the common good. Public inquiry
165
enables victims’ circumstances to be taken into account and thus implies that, by
highlighting their sacrifice, their circumstances may be made more acceptable.
Dismissing the objections of those opposed to policies as mere manifestations
of the ‘Nimby syndrome’ is therefore an effective means of discrediting actors
reluctant to embrace the rules of the game.
These mechanisms indirectly enable state authorities to assess the public’s
mobilisation on a given issue, providing a variety of potential options to decision-
makers who may or may not use them. Put differently, even in cases where
participatory mechanisms allow new policy knowledge to be created, there can
be no guarantee that this knowledge will be incorporated into policy-making.
Studies on environmental justice have shown how some locations are chosen
based on particularly detailed knowledge of the likelihood of protest depending
on the socioeconomic characteristics of the residents concerned. In an attempt to
intensify monitoring of the population, studies may also be used to more effectively
identify potential opponents. Producing different forms of lay knowledge is thus
insufficient; indeed, the knowledge must also be integrated into the decision-
making process and the conditions under which it will be ‘taken into account’
must be clearly defined.
What does ‘taking into account’ mean? Highlighting the democratic nature of the
decision-making process
Studies on policy analysis and policy process study have shown relatively little interest
in the knowledge produced via public participation mechanisms. This in itself is
revealing. Indeed, the majority of these studies have found that this knowledge
plays only a minor role in policy-making (Papadopoulos and Warin, 2007).
Contrary to what ‘participation studies’ – which are occasionally normative – may
suggest (Blatrix, 2012), it is questionable whether they have any real impact on
the content of policies (Blatrix, 2010).
While public participation mechanisms help change the power relations between
actors or give greater visibility to certain kinds of knowledge such as expert
knowledge, these mechanisms do not provide a body of knowledge capable of
challenging conventional policy-making processes. The use of deliberative mini-
publics in France is significant: while these publics are valued as an innovative
democratic experience (Gourgues, 2010), they generally represent isolated and
one-off operations that hardly challenge the decision-making process. Moreover,
outcomes (opinions delivered, recommendations, diagnostics) are integrated into
policy analysis only when they are in line with pre-established and dominant
knowledge (Barbier et al, 2009). Broadly speaking, the role of participatory
mechanisms in policy analysis is guided by a rather restrictive framework and by
highly specific expectations. As Archon Fung (2007) argues, public participation
is often analysed in terms of policy failures which participation is expected to
address. Public participation may thus allow citizens to develop their preferences;
it may also monitor changes in preferences, strengthen the accountability of
166
decision-makers and improve the policy implementation process. This therefore

explains why the mechanisms that do not fulfil their assigned functions, or those
that go beyond expectations, are unlikely to be integrated into the decision-
making process.
This ambiguity does not only result from the restrictive framework that
constrains public participation mechanisms. Applying a gender perspective to
participatory mechanisms, Marion Paoletti and Sandrine Rui (2015) argue that
participation mechanisms could prevent, in a sophisticated and co-produced way,
the emergence of a feminist critical discourse on policies. On the one hand,
women attend participatory mechanisms and should, logically, produce such a
critique, insofar as their presence supposes that a gender perspective is ‘represented’
in decision-making processes. On the other hand, however, the production of
feminist critical policy analysis is often avoided by participants themselves. In the
case of males, a universal framing of citizenship and public debate is opposed to
gendered perspectives. By contrast, women face a so-called ‘Mary Wollstonecraft
dilemma’ (Pateman, 1988) and are caught between claims for equality and/or for
difference. Echoing gender policy studies in other national contexts, applying
a gender perspective to participation mechanisms in France highlights the need
to distinguish between gender inclusiveness in public participation and the
development of feminist critical policy analysis within policy-making. It also
provides an insightful explication of why a large number of feminist movements
still prefer counterpublic strategies (Fraser, 1990), which ensure higher levels
of autonomy and critically empowering women, rather than participating to
top-down public participation mechanisms. Although single-sex participatory
mechanisms do not guarantee a better ability to shape policy outcomes (Sa
Vilas Boas, 2016), it nevertheless facilitates the emergence of feminist counter-
hegemonic discourses.
Decisive moments for dominant policy analysis and increasing the visibility of
alternative frameworks
Individuals may participate in public inquiry or public debate in an attempt

to evaluate the soundness of the knowledge on which projects are based. By
occasionally undertaking a very thorough analysis of the conditions of production
of ‘official expertise’, by demanding or conducting counter-expertise, and by
proposing alternative solutions, participants not only subject state knowledge to
the test of truth but also assess its soundness. They test the relevance of policy
analysis that lies behind projects under discussion (for instance, concerning
transport infrastructure or traffic forecasting [Doridot, 2007]) as they force
project leaders to disclose their assumptions and methods, and even their beliefs
and perceptions of society. The expertise behind the projects may thus emerge
as faulty, and it may prove to be outdated and/or to be more closely associated
with the preoccupations of a specific group of actors rather than the interests
of the public at large (irrespective of whether this concerns state or non-state
167
actors such as major sports federations or professional organisations) and local

authorities. The recent conflict around ‘unnecessary imposed mega projects’
provides a number of examples to illustrate this point.7
Provided that these mechanisms are recognised as such when they are mobilised,
they could provide spaces enabling different policy analyses to be compared by
undermining the dominant expertise and increasing the visibility of the policy
analysis undertaken by objective challengers who are rarely granted access to the
public space. Ultimately, these mechanisms bring existing arguments and world
views to the fore, even if this means re-establishing the political dimension of
competing social projects. Irrespective of whether they involve public inquiries,
committees or public debates, participatory mechanisms can become privileged
spaces for highlighting alternative frameworks. These alternative frameworks go
beyond simply providing counter-expertise as they can pit different definitions
of ‘public interest’ against one another. Actors are thus encouraged to commit
to social learning which forces them to re-examine their initial preferences.
However, the impact of this process on actors’ perceptions and strategies must
not be overestimated (Muro and Jeffrey, 2008).
Public participation can thus go beyond its initially intended framework and
directly influence policy analysis. As evidenced by the example of the burial of
nuclear waste (Barthe, 2002), the production of expert knowledge can elude
policymakers. As a result, power relations favouring some form of counter-
expertise or the dominant view may emerge and obstruct the decision-making
process. In the case of nuclear waste mentioned above, the scope of the public
problem evolved during the public inquiry, following a struggle that highlighted
competing arguments. A number of public inquiries have given rise to large-
scale mobilisations and thus enabled activist networks to challenge the directions
in which policies have developed: in the transport sector, the primacy of the
motorway network at the expense of the railway was denounced by opponents,
at a time when environmental concerns were beginning to be heard. The same
issue was raised in relation to the development of high-speed trains at the expense
of local railway services.
The circulation of the policy process analysis
Whatever their degree of control, participatory mechanisms bring different

viewpoints to the fore and help give them a voice. They go well beyond policy
analysis and analyse the policy-making process itself. Put differently, public
participation mechanisms provide knowledge on policies that drive actors’
actions as well as on how government and politics function. The creation of the
aforementioned French public debate procedure is worth mentioning. Towards the
end of the 1980s, a public inquiry was established to examine a new high-speed
train project (TGV Méditerranée). It was during this process that critics of public
inquiry and of the decision-making process itself voiced strong criticism. The lack
of transparency in the decision-making process was denounced because the major
168
infrastructure projects were seen as imposed by the state. While the opponents
did not dispute the choice of location, they challenged the very relevance of the
projects. Following this conflict, a procedural concession to involve opponents
was made. Discussions were held on how the concept of public utility might be
reformed and a new procedure referred to as ‘public debate’ was created. This
procedure was to be carried out under the auspices of an independent commission
entitled the ‘National Public Debate Commission’.
Analysing how these mechanisms function thus draws our attention to another
important characteristic of public participation: while public participation
produces knowledge on public policy (policy analysis), it invariably results in an
analysis of the decision-making process itself (policy process analysis). Criticism
has often been levelled against the decision-making process during the moments
of mobilisation previously mentioned: decisions were hardly based on sound
and reliable policy studies, choices lacked transparency and were dependent on
political calculations, and the public decision-making process favoured specific
questionable interests. Far from unanimously supporting the organisation of
public inquiries and debates, participants took advantage of public participation
mechanisms to explicitly challenge the decision-making mechanisms and the
legitimacy of both dominant knowledge and decision-makers, forcing them to
justify themselves. Analysing how public debates function, Sandrine Rui (2004)
has shown that the decision-making process itself often fuels controversy. French
public inquiry thus has a rather intriguing nature. While participatory mechanisms
are set up to address the social criticism levelled against the decision-making
process (policy process analysis), the manner in which these mechanisms function
continues to reveal that these problems remain persistent and that the decision-
making process never really changes. Recent examples of violent conflicts around
major French development projects, including the zones to be defended (zones
à défendre [ZAD]) of Notre-Dame-des-Landes [airport project] and Sivens [dam
project], are strongly reminiscent of environmental conflicts in the 1970s and
1980s. These examples show that participatory mechanisms do not systematically
lead to a greater acceptability of proposed projects.
Conclusion
Analysed in terms of policy analysis, participatory mechanisms such as public
inquiries or committees lie at the heart of a certain degree of tension: while they
are rarely set up to produce original knowledge on public policies, the mobilisation
they favour may occasionally reorganise the long-held conventional knowledge
and invoke criticism of the decision-making process itself. As evidenced by the
French example, it thus seems crucial to analyse public participation from a
broader perspective that focuses on the conventional forms of policy-making and
takes into account representative, administrative, economic and scientific elites.
The design, implementation and functioning of participatory mechanisms are
closely associated with shifts in the power relations within and beyond the circle
169
of influence of the institutional actors responsible for policy-making. These power

relations largely determine what, if anything, public participation mechanisms
might change in terms of policy analysis.
Notes
1 See Chapter Thirteen.
2 See Chapter Fourteen.
3 According to article L1223-1 of the Environmental Code, the objective of public inquiry is to
inform the public and allow them to participate in decision-making in issues likely to affect the
environment. All collected information is taken into account by both the project supervisor
and the decision-makers.
4 An analysis of parliamentary sessions at the National Assembly in 1975-1976 shows that public
participation is viewed as a guarantee for the quality of impact assessment (Nungesser, 1976).
The draft decree communicated to the managing committee specified that the project manager
was solely responsible for conducting the impact assessment or, alternatively, for commissioning
someone to conduct it, and he/she alone was responsible for drawing the consequences from
the assessment’s conclusions. As such, the quality of the assessment was not guaranteed and
managers could even view impact assessments as a mere formality: ‘Indeed, nothing can
guarantee the quality and reliability of the study commissioned by the project manager. To
meet the legal requirements, he/she needs only give a reasoned opinion on a study even if it
provides no guarantees. The project manager is thus free to harm the environment as he/she
wants’ (Nungesser, 1976). Consequently, project managers are required by the Commission
responsible for studying the draft law to make the environmental impact assessment public:
‘Everyone can thus be informed of the consequences of the planned works and the mere fact that
the assessment is made public should deter its authors from being too (sic) sloppy’ (Nungesser,
1976).
5 There has been a recent decrease (to approximately 5,000 projects) following the recent reforms
seeking to simplify environmental law.
6 While the Solidarity and Urban Renewal Act (SRU) enabled these committees to be set up,
regional councils were not legally bound.
7 We can briefly cite the project of the Sivens dam constructed across the Tescou River and the
project seeking to expand the Roland Garros stadium.
References
Akrich, M., Barthe, Y., Rémy, C. (2010) Sur la piste environnementale: Menaces
sanitaires et mobilisations profanes, Paris: Presses de l’ École des Mines.
Barbier, M. (2013) ‘Pour une approche pragmatique, écologique et politique de
l’expertise’, Revue d’anthropologie des connaissances 7(1), 1–23.
Barbier, R., Bedu, C., Buclet, N. (2009), ‘Portée et limites du dispositif “jury
citoyen”: Réflexions à partir du cas de Saint-Brieuc’, Politix 2(86), 189–207.
Barnes, M., Newman, J., Sullivan, H. (eds) (2007) Power, Participation and Political
Renewal: Case Studies in Public Participation, Bristol: Policy Press.
Barthe, Y. (2013) ‘Rendre discutable: Le traitement politique d’un héritage
technologique’, Politix 15(57), 57–78.
Blatrix, C. (2010) ‘Démocratie locale et débat public’, in O. Borraz, V. Guiraudon
(eds) Politiques publiques 2, pp. 213–42, Paris: Presses de Sciences Po.
Blatrix, C. (2012) ‘Des sciences de la participation: Paysage participatif et marché
des biens savants en France’, Quaderni 79, 59–80.
170
Bongrand, P., Gervais, J., Payre, R. (2012) ‘Les savoirs de gouvernement à la

frontière entre “administration” et “politique”’, Gouvernement et action publique
1(4), 9–20.
Callon, M. (1998) ‘Des différentes formes de démocratie technique’, Annales des
mines - Responsabilité et environnement 9, 63-73.
Callon, M., Lascoumes, P., Barthe, Y. (2013) Acting in an uncertain world, Boston,
MA: MIT Press.
Cartwright, N. (2007) Evidence-based Policy: Where is our Theory of Evidence?,
London: London School of Economics and Political Science (LSE) and San
Diego, CA: University of California, San Diego (UCSD).
Chilvers, J. (2008) ‘Environmental risk, uncertainty, and participation: Mapping an
emergent epistemic community’, Environment and Planning A 40(12), 2990–3008.
Corburn, J. (2005) Street Science: Community Knowledge and Environmental Health
Justice, Cambridge, MA: MIT Press.
Devlin, J., Yap, N. (2008) ‘Contentious politics in environmental assessment:
Blocked projects and winning coalitions’, Impact Assessment and Project Appraisal
26(1), 17–27.
Doridot, F. (2007) ‘Débattre publiquement autour des chiffres: Quelques enjeux
de la controverse sur les flux dans le débat LAALB’, in M. Revel, C. Blatrix,
L. Blondiaux, J-M. Fourniau, B. Hériard Dubreuil, R. Lefebvre (eds) Le débat
public: Une expérience française de démocratie participative, Paris: La Découverte.
Fiorino, D.J. (1990) ‘Citizen participation and environmental risk: A survey of
institutional mechanisms’, Science, Technology and Humans Values 15(2), 226–43.
Fischer, F. (2000) Citizens, Experts, and the Environment: The Politics of Local
Knowledge, Durham, NC: Duke University Press.
Fischer, F. (2012) ‘Participatory governance: From theory to practices’, in L.-F.
David (ed) Oxford Handbook of Governance, Oxford: Oxford University Press.
Fraser, N. (1990) ‘Rethinking the public sphere: A contribution to the critique
of actually existing democracy’, Social Text 25/26, 56–80.
Fung, A. (2007) ‘Democracy and the policy process’, in M. Moran, M. Rein, R.
Goodin (eds) Oxford Handbook of Public Policy, pp. 669–85, New York, Oxford:
Fung, A., Wright, E.O. (eds) (2003) Deepening democracy: Institutional innovations
in empowered participatory governance, London: Verso.
Goodin, R.E. (2008) Innovating Democracy: Democratic Theory after the Deliberative
Turn, Oxford: Oxford University Press.
Gourgues, G. (2010) ‘Studying the region as a space for democracy: A sociological
approach of regional politics’, Regional and Federal Studies 20(3), 353–69.
Gourgues, G. (2012) ‘Des dispositifs participatifs aux politiques de la participation.
L’exemple des conseils régionaux français’, Participations 1(2), 30–52.
Hisschemöller, M., Hoppe, R., Dunn, W.N., Ravetz, J.R. (eds) (2001)
‘Knowledge, power, and participation in environmental policy analysis: An
introduction’, Policy Studies Review Annual 12, 1–28.
171
Kessler, M.-C. (1999) ‘Pourquoi n’y a-t-il pas de corps de l’environnement?’,

in P. Lascoumes (ed) Instituer l’environnement. Vingt-cinq ans d’administration de
l’environnement, pp. 71–85, Paris: L’Harmattan.
Lascoumes, P. (1994) L’éco-pouvoir: Environnements et politiques, Paris: La
Découverte.
Lascoumes, P. (2002) ‘L’expertise, de la recherche d’une action rationnelle à la
démocratisation des connaissances et des choix’, Revue française d’administration
publique 103(3), 369–77.
Le Clainche, M. (2011) ‘L’administration consultative, élément constitutif ou
substitut de la démocratie administrative ?’, Revue Française d’Administration
Publique, 137–8(1), 328–48.
Lepage, C. (1998) On ne peut rien faire, Madame le Ministre…, Paris: Albin Michel.
Lowndes, V., Pratchett, L., Stoker, G. (2001) ‘Trends in public participation:
Part 1 – Local government perspectives’, Public Administration, 79(1), 205–22.
Mazoyer, H. (2012) ‘La construction du rôle d’ingénieur-économiste au ministère
des Transports: Conseiller le politique, résister au comptable et discipliner le
technician (1958–1966)’, Gouvernement et action publique 4(4), 9–20.
Muller, P. (1984) Le technocrate et le paysan, Paris: Les Éditions ouvrières.
Muro, M., Jeffrey, P. (2008) ‘A critical review of the theory and application of
social learning in participatory natural resource management processes’, Journal
of Environmental Planning and Management 51(3), 325–44.
Nez, H., Sintomer, Y. (2013) ‘Qualifier les savoirs citoyens dans l’urbanisme
participatif ’, in A. Deboulet, H. Nez (eds) Savoirs citoyens et démocratie urbaine,
Rennes: Presses universitaires de Rennes.
Nungesser, M. (1976) Rapport au nom de la commission de la production et des échanges,
National Assembly, 2372, Paris: Assemblée Nationale.
Paoletti, M., Rui, S. (2015) ‘La démocratie participative a-t-elle un sexe?’,
Participations 2(12), 5–29.
Papadopoulos, Y., Warin, P. (2007) ‘Are innovative, participatory and deliberative
procedures in policy-making democratic and effective?’, European Journal of
Political Research 46(4), 445–72.
Pateman, C. (1988) The Sexual Contract, Cambridge: Polity Press.
Pratchett, L. (1999) ‘New fashions in public participation: Towards greater
democracy?’, Parliamentary Affairs 52(4), 616–33.
Rui, S. (2004) Les citoyens face à l’action publique, Paris: Armand Colin.
Sa Vilas Boas, M.-H. (2016) ‘The social anchoring of representation: Becoming
a spokesperson in municipal women’s conferences in Recife, Brazil’, Revue
Schäfer, A., Streeck, W. (eds) (2013) Politics in the age of austerity, Cambridge:
Polity Press.
Smith, G. (2009) Democratic Innovations: Designing Institutions for Citizens
Participation, Cambridge: Cambridge University Press.
Spanou, C. (1991) Fonctionnaires et militants: L’administration et les nouveaux
mouvements sociaux, Paris: L’Harmattan.
172
Suaud, C. (1984) ‘Le mythe de la base: Les Etats Généraux du développement

agricole et la production d’une parole paysanne’, Actes de la recherche en sciences
sociales 52–3(1), 56–79.
Sunstein, C.R. (2006) Infotopia: How Many Minds Produce Knowledge, New York:
Weber, Y. (1968) L’administration consultative, Paris: Librairie générale de droit et
de jurisprudence.
Wynne, B. (1996) ‘May the sheep safely graze? A reflexive view of the expert–lay
knowledge divide’, in S. Lash, B. Szerszynski, B. Wynne (eds) Risk, Environment
and Modernity, pp. 44–83, London: Sage.
173
TEN
Management consultants
as policy actors
Julie Gervais and Frédéric Pierru
In the literature related to corporate participation in public policies, the case

of France is often considered as an exception. French so-called resistance to
consultants is often put down to the organisation of its senior civil service around
Grands Corps. It is the very existence of these prestigious bodies deemed to show
resistance to any kind of competitors coming from the private sector which would
explain the fact that corporate actors are less present in the public sector than in
other countries (Saint-Martin, 2000a; 2000b). This thesis conceals the growth
of ‘consultocracy’ (Hood and Jackson, 1991, 224) in France since the beginning
of the 2000s, but it also artificially opposes two groups: corporate actors on the
one side and public ones on the other, which are both supposed to protect the
price and the legitimacy of their own expertise. A close sociological study of their
training, trajectories and careers shows that they are much more intermingled and
similar than the common picture seems to indicate. Being in a sociological position
of homology, many consultants and high civil servants in France have common
interests, speak alike, share the same values and follow similar careers (Gervais,
2012). This paper summarises the main steps which contributed to consultants’
increasing deployment in French public policies and seeks to describe the type
of service they aim to deliver, as well as what decision-makers may consider as
their ‘added-value’. It argues that a critical part of their increasing deployment
within French public policies and administrative reforms relies on their legitimising
effects over change. By doing so, it highlights the national variations at stake in
terms of their involvement and the extent to which they have an impact on the
French public sector.
The consultant: the new state expert icon

French nineteenth-century experts belonged to a clearly defined and bounded
world, and were closely linked to academia and the scientific field. They
derived their social recognition, credibility and legitimacy from their scientific
knowledge (Berrebi-Hoffmann and Lallement, 2009). In sharp contrast with these
nineteenth-century characters, today’s experts work for corporate firms, devise
performance targets and set efficiency savings programmes. This shift occurred
during the late twentieth-century in France, when state reform expertise moved
from the field of intellectuals and academics to that of consultants and think
tanks (Berrebi-Hoffmann and Grémion, 2009). France was first characterised
175
by its internal expertise capacity via the state-controlled training of experts and
controllers belonging to the senior civil service, organised into unified and quite
powerful bodies called Grands Corps. In the 1970s, large consulting firms started
getting involved in public sector ‘modernisation’ process. For example, the
corporate transformation of the French postal service was prepared by confidential
reports from McKinsey & Co (Vezinat, 2015). During the heyday of French
consulting firms, in the 1980s, technical Grand Corps and some ‘intellectuals’
closely worked together. They used organisational knowledge promoted by
anti-bureaucratic thinkers such as Michel Crozier, himself closely related to right-
wing managerial circles and governmental boards and commissions (Magnin,
2010). Specialised in organisational issues, French consultancy firms acted as
mediators between Anglo-Saxon managerial culture and administrative elites’
knowledge (Berrebi-Hoffmann and Grémion, 2009). Hence until the 1990s,
state expertise was a matter for high civil servants, political elites, some business
leaders, along with a few carefully selected academics, intellectuals, journalists
and union leaders gathered in closed clubs. They soon will be overthrown by
big Anglo-Saxon consultancy firms which underwent major economic growth
in France, especially as a result of restructuration programmes accompanying
French public companies in their privatisation process during the 1990s. From
the 1990s onwards, alongside the expansion of new public management (NPM)
ideology on the continent, big consultancy firms started investing the flourishing
market of French general government by intensifying their contacts with senior
civil servants (Bezes, 2012), transferring their engineering knowledge from big
companies to state related matters, and by hiring high civil servants and graduates
from French Grandes Écoles. After undergoing a period of concentration, these
big companies opened up ‘public sector and government’ departments, first
focused on data processing, then on audit and public accounting reforms, and
finally on ‘state reorganisations and de-bureaucratisation’ (Berrebi-Hoffmann and
Grémion, 2009, 56). These companies now provide a large variety of services,
from IT infrastructures, benchmarking databases, change management projects,
cost and quality service improvements to organisational design, risk assessment
studies, cost-cutting reforms and outsourcing strategies.
If the Commissariat à la réforme de l’État temporarily resorted to consultants
in the late 1990s (Bezes, 2012, 30), it is really at the beginning of the 2000s
that contacts between consulting companies and senior civil servants started
developing on a more regular basis. Hence two major state reforms are symbolic
of the intensification and the widespread diffusion of NPM standards in French
administration. First, the Constitutional bylaw of 2001 on budget acts (also known
as LOLF in French), enforced in 2006, introduced new rules for preparing and
implementing the state budget by moving from a resource-based to a results-
based approach. It set the institutional and legal conditions for public spending to
become outcome-oriented with performance indicators, justification from the first
euro spent, rendering of accounts, and results evaluation. The LOLF mobilised
consultancy firms such as McKinsey-Accenture, Capgemini-Boston Consulting
176
Management consultants as policy actors
Group or Ernst & Young which played a major role in the 2006 ‘modernisation
audits’ which put together high civil servants and consultants in order to closely
examine public policies in a budgetary-cuts approach. Launched in 2007 and
inspired by the Canadian Program Review, the French General Review of Public
Policies (RGPP) placed strategy consultants ‘at the heart of the state’, assigning
them ‘a role unheard of in the French context’ (Bezes, 2012, 33). The RGPP
initiated both a quantitative and a qualitative shift in the involvement of consulting
firms within the French state. It opened up a very specific configuration of state
reform introducing two major changes in comparison with previous reforms. On
the one hand, ‘state reform’-related matters fell within the Ministry of Budget’s
remit, symbolising a turning point to a more budget-based approach. On the
other hand, as was the case with its Canadian counterpart, the RGPP was highly
politicised. It was at the heart of Sarkozy’s campaign for the 2007 presidential
election as he framed it as a symbol of his political voluntarism and leadership.
Once Sarkozy was elected, the first steps of the RGPP were launched in May–June
2007, with an early involvement of Capgemini. The Directorate-General for the
Modernisation of the State (Direction générale de la Modernisation de l’État [DGME]),
an interministerial body responsible for coordinating the transformation of French
government, was in charge of organising auditing teams which paired high civil
servants and consultants from major multinational firms. By favouring the latter,
the DGME excluded smaller consulting firms from the deal and gave priority
to standardised international solutions (Bezes, 2012, 32) designed by McKinsey
Accenture, Capgemini-Boston Consulting Group and Ernst and Young. The
RGPP thus enabled strategy consultants to make an entrance at the heart of the
French state. At the articulation between French administration and the corporate
world, the DGME, led by François-Daniel Migeon, a senior civil servant who
worked as an associate partner at McKinsey company for eight years, was staffed
with consultants on short-term contracts coming from major companies such
as Boston Consulting Group, Bain & Company, McKinsey, Ernst & Young,
PriceWaterhouseCoopers, Deloitte, and so on. Between 2007 and 2011, the
budget devoted to auditing services and modernisation impact studies amounted
to €111 million.
If these sequences represent major trends, many public policies have their specific
temporality and logic (Boussard et al, 2015). For example, it was shown that French
police reforms were less radical than Anglo-Australian ones because external
consultants and experts were less involved in their design (de Maillard, 2009).
However, ad hoc policy decisions on ‘urban safety’ in particular contributed to
the consulting market boom in the mid-1990s. By requiring that local authorities
carry out ‘local security diagnostics’, French executives facilitated the structuring
of a consultancy market dominated by two large firms (Bonelli, 2009). In the
hospital sector, reforms from the mid-1980s also contributed to the emergence
of small consultancy firms, often founded by former health professionals and
hospital managers. The gradual deployment of NPM in the 1990s did not create
a consultancy market ex nihilo; rather it played a part in its restructuring (Pierru,
177
2012). Soon though, small firms were crowded out by major transnational firms
entering the market while the state reform went into ‘industrial mode’ (Bezes,
2012). In other words, every public policy area reflects the dominant trends in
its own history and its institutional configuration.
Felicity conditions for resorting to consultants

The circulation between the private and the public spheres helps to understand
how a hybrid elite of general government works hand in hand in the development
and diffusion of NPM ideas within the French state. The ideological impregnation
which turns the corporate system into a model and NPM doctrine as an obvious
and imperious necessity also finds its roots in the evolution of French high civil
servants’ training (Gervais, 2007) and in civil service competitive examinations’
reforms such as that of hospital managers which subsequently changed the
social backgrounds of its candidates, now more willing to ‘take on a rewarding
managerial role’ (Pierru, 2012, 37). However, the control which consultants have
over public sector organisations they are paid to transform is not only due to their
social proximity to political and administrative elites. In the case of gender experts
and consultants promoting diversity, equal opportunities and environmental
protection within the public sector, a market which has significantly grown
since the 2000s in France, specialists circulate back and forth from the worlds of
academia, social movements and expertise. Consultants can also find allies lower
in the hierarchy. In the case of the privatisation of electricity (EDF), consultants
found resonance among individuals in non-managerial positions who had spent
their entire career in the public company: those forced to take circuitous routes in
order to get promoted saw the collaboration with consultants as an opportunity
to show their technical expertise in accordance with their ‘public service spirit’
(Thine, 2012). Similarly, in hospitals, consultants met the interest of professionals
working in dominated medical areas (Pierru, 2013) or positioned at the margins
of their organisation (Belorgey, 2010). Thus consultants can rely both on their
proximity to political and administrative elites and their one-off alliances with
collective and individual actors dominant in their professional positions.
Beyond the mapping of alliances which created work and contract opportunities,
one has to pay attention to what consultants actually provide in order to understand
the conditions in which executives resort to them. Consultants’ contribution often
comes down to a methodological input (Boni-Le Goff, 2012). Their added value
stems from the gradual development of standard tools transferable from one sector
to another, and not from any specific knowledge regarding the subject or area
they are meant to reorganise – this knowledge often being non-existent at the
start of their missions. Their tools are of four types: work organisation in project
mode; methods of simplifying data processing and analysis production; methods
of collaborative work (animation methods, for example); and presentation of
the results, whereby PowerPoint supplants the traditional administrative report.
These skills and tools are not generally mastered by administrators. In the 2009
178
case of the merger of state services and health insurance in French regional health
agencies (ARS), a ‘project group’ was set up following the implementation of
the law. This group included a handful of senior officials, acquainted with the ins
and outs of health administration, but also many junior consultants who were not
familiar with the area. Three reasons explain the pervasiveness of consultancy in
this case. First, the merger process bears the mark of the RGPP which involved a
qualitative leap in the use of consultants by the state. Second, the ‘project group’
struggled to recruit senior officials as they were reluctant to take on temporary
positions; and third, officials poorly mastered the skills and techniques of project
management. Hence interviews conducted with officials show their ambivalence
toward consultants: they often lament consultants’ ignorance of the subjects at stake
but they concede that the latter provided resources which French administration
internally lacked. Even when they were sceptical, these officials considered that if
it hadn’t been for the consultants’ work, the reform could not have been carried
out within the very short timeframe imposed by political executives.
Finally, the success of consultancy lies in the legitimising effects it has on public
policies and administrative reforms. Management consultants admit themselves
‘sugaring the pill for the staff’ (Pierru, 2012, 42), helping executives implement
their agendas, or gathering information and material in order to justify a decision
already made. Even gender equality consultants, who work on translating feminist
theories to practitioners (Bustelo et al, 2016, 13) and seek to get this information
‘incorporated into formally adopted rules, statutes, actions and/or decrees of the
organisation in question’ (Hoard, 2015, 3), cannot always escape the ‘smothering
of feminist knowledge’ (Davids and Eerdewijk, 2016) and its colonisation by
neoliberalism (Rottenberg, 2014). Often, decision-makers are after a ‘linguistic
paraphernalia’ (Sorignet, 2012, 26) which would be authoritative insofar as it
comes from the private sector, sanctioned as the legitimate model to be followed.
The very mention of the consulting firm’s label as the source of the diagnosis
establishes the authority of the decision and tends to make the arbitrary nature of
the recommendations acceptable (Gantenbein, 1993). The service of management
consultants may also be helpful insofar as it provides an external point of view
so that executives seeking blame-avoidance can distance and protect themselves
from possible controversies. It furthermore enables decision-makers to bypass
potential resistance from public agents. Their externality and the temporality
of their interventions reduce the potential for the formation of an autonomous
group which would gather support and oppose ‘change’ from within. According
to NPM’s anti-bureaucratic ideology and the distrust of civil servants it promotes,
resorting to consultants acts as a check on any hint of resistance from the primarily
concerned staff. Hence far from the claim of neutral and objective expertise, one
can see management consultants’ intervention as inherently political in the way that
it acts as a useful instrument in order to overcome oppositions, weaken internal
positions and make people accept controversial reforms (Poupeau et al, 2012).
In the end, research conducted on various public policies show that the structure
of consultancy markets, the role and degree of penetration of consultants in
179
decision-making processes, and their impact on final results and outputs widely
vary. Even after the 2000s, consultants are often not in a situation of omnipotence
(Poupeau et al, 2012). From this point of view, the ‘cognitive tutelage’ exercised
by the few specialists in financial engineering over public policymakers about
the private finance initiative (PFI) (Deffontaine, 2012) cannot be compared
to the weak position of small consulting firms intervening in the ‘strategy and
urban projects’ market (Linossier, 2012). In one case, consultants coming from
major companies who master highly technical financial instruments are the only
ones who can get into the black box of contract tendering. Policymakers are
content with a few indicators and ratings on the various projects tendered. The
asymmetry of skills between them is difficult to reduce. The first reason is because
of a lack of financial and internal resources in the public sector. The second is
because the circulation at work plays against such a trend: civil servants who could
receive specific training would immediately get poached by a private company
offering an incomparably attractive salary (Deffontaine, 2012). In the other case,
the situation is dominated by officials for the consultancy market is fragmented,
unclear, unstable and economically weak. In this case, consultants are only
secondary participants in a market which is the preserve of public organisations.
Consulting firms working for this atypical trade have no choice but to develop
business strategies to bring out their professional network or to diversify their
skills and professional experiences. Consultants’ careers in town and country
planning evolve at the articulation of several professional circles, between public
and private sectors, political and technical worlds, academic and practical fields.
Resources mobilised also vary depending on configurations. In the first
case, the technical monopoly of consultants is a key resource in relations with
policymakers. In the second, the consultants have ‘niche strategies’ and they rely
on relationships, particularly among their political networks. Finally, the border
between consultancy and final decision is mobile, according to the balance of
power between the different members of the consulting relationship. Hence the
thesis of ‘consultocracy’ requires careful handling. We should not only consider
the few striking reform cases in which consultants from multinational consulting
companies played a leading role. Here as elsewhere, the researcher must pay
attention to the particularities of public policy sectors, their history and their
institutional configuration.
The impact of consultancy

If some authors consider that management consultants’ implication in the reform
of the state has ‘powerful effects on the content and the forms of reform policies’
(Bezes, 2012, 35; Saint-Martin, 2000a), the great paradox which characterises
consultants’ work, according to others, is the impossibility of assessing their
interventions and their impact in the name of their prescriptions’ autonomy
(Thine, 2006). Because consultants strive to disappear behind their principals,
their activities remain in the shadows, especially during the development phase
180
of reforms. Then the instruments they develop constitute the bulk of their firms’
capital: the secret is jealously guarded and often sheltered from researchers.
However, some social scientists managed to get hired as consultants in public
agencies, hence getting the opportunity to watch consultants at work, as in the
case of Nicolas Belorgey who studied the deployment of NPM in French public
hospitals (Belorgey, 2010). Critical to the consultant’s impact will then be her
or his position in the division of labour reform which can vary greatly from one
public sector to another. In other areas, a limited amount of research shows how
consultants directly inspired the content of reforms they led. Thus, the few firms
specialised in lean management directly affected the streamlining of administrative
procedures, particularly in prefectures, under the RGPP sequence of French state
reform process (Bezes, 2012).
While it is often practically ‘impossible to determine whether a consultancy
mission has succeeded or failed’ (Henry, 1992, 38), the attempts to control the
controllers seem nevertheless to have led to an ever-expanding auditing machinery,
multiplying assessment levels with a proliferation of indicators in a systemic and
profitable logic (Berrebi-Hoffmann and Lallement, 2009, 8–11) with a clearly
perceptible impact: the self-reproduction of the consultancy market. In their search
for profit, management consultants create problems which they will be paid to fix,
producing a self-maintained consulting inflationary spiral. It is the case with the
development of PFI whereby large consulting firms, in spite of the accumulated
failures of this mode of public investments, encourage the development of the
market by ‘educating’ their public customers (Deffontaine, 2012) – what they
also call ‘market education’ in order to ‘open up officials’ minds’, ‘create mature
customers’ and ‘raise executives’ awareness’ (Gervais, 2012, 19).
Existing research also shows the impact of consultancy standardisation on public
sector reforms and public policies. Time pressure, the availability of business
models from the private sector to be made standard, as well as the deliberate
ignorance or lack of knowledge of public sector’s constraints and specific
values lead management consultants to safely choose the route of imitation
and ready-made solutions. In the case of the French public radio and television
reform, consultants ‘eradicate[d] any organisational memory’ by adopting a
‘de-historicising approach’ (Sorignet, 2012, 28) and drastically framing what
was thinkable and what was not, what was relevant and what was not, what was
feasible and what was not. In this case, the appointment of previously private-
sector managers in key positions in the public radio and television sector opened
up new opportunities for consultants. The idea that the reform was about putting
an end to cumbersome bureaucracy and unnecessary administrative procedures
was constantly reinforced both by public managers and private consultants.
‘Public ways’ of dealing with reforms were systematically disparaged in regards to
a supposedly flexible, effective and efficient ‘private-sector approach’ to change.
By resorting to an obscure and technical wording, incomprehensible to non-
specialists, consultants managed to neutralise potential criticisms from employees.
The latter were left totally confused, incapable of levelling any criticism as they just
181
‘didn’t understand a damn thing!’ (Sorignet, 2012). It was even more difficult to
accept that it was the very definition of their job as journalists which was at stake.
Hence, the employment of private consultants by the new managerial team aimed
to naturalise the reform and avoid ‒ or ‘exernalise’ ‒ responsibility for any blame
while, at the same time, making its content very difficult to grasp by those who
were directly affected by it. In this way, consultants exert a real ‘symbolic violence’
over employees, while the plans and procedures they implement deeply transform
the latter’s daily professional activities. This double dispossession, both lexical and
practical, undergone by civil servants and employees of the public sector in regards
to the reforms they are subjected to is at work in other fields of state reform. In
French hospitals for example, the neo-managerial language spread by consultants
and used by the new leadership who now run hospital complexes and their staff
doesn’t fit the work realities of clinicians. The latter find this mismatch difficult
to express and lack the categories and standards through which to oppose this
neo-managerialism. How indeed can one counter organisational transformations
which claim to adhere to ‘efficiency’ and ‘performance’, invoke ‘healthcare quality’
and refer to the alleged objectivity of managerial quantification? The healthcare
elite is discredited and denigrated as the representative of the ‘old world’ attached
to its privileges while ‘the modern’ managers associate themselves with apparently
irrefutable and objective standards of ‘performance’, ‘efficiency’, ‘accountability’
and ‘healthcare quality and security’. These standards are further fortified by being
embedded in larger discourses of economic growth and democracy.
The public sector ends up with hastily tacked-on grid and schemes, often totally
‘unsuited to administrative realities’ (Pierru, 2012, 45). In the French health sector,
consultants − who increasingly come from large consultancy firms – totally lack
‘organisational memory’ or the very understanding of professional identities and
practices of the sector. The exteriority of management consultants’ intervention
allows them to keep staff at bay and ignore the people primarily targeted by the
measures they prescribe. The marginalisation or exclusion of civil servants and the
absence of recognition of their skills, competencies and indigenous knowledge
bears a strong link to the impact of management consultant’s intervention on
public agents’ work, stress and motivation. In the case of French hospitals’ reforms,
consultants’ industrial vision of medicine run counter to the autonomy of health
professionals. Closely associated with the development of HPST (relating to
hospitals, patients, health and territories [hôpital, patients, santé, territoires]), as part
of the RGPP, they implemented reorganisation schemes that did not take into
account the historical institutional rivalry between the French state and French
public welfare system (Sécurité sociale). In line with the Trojan horse strategy,
however, their implication in designing the law allowed them to conquer new
markets: they are now closely associated with hospital reorganisations and the
certification of hospitals’ accounts. At this stage, they contribute to the trivialisation
of public service delivery: they apply blunt tools and tack on worldviews that are
largely unaware of the hospital sector’s specificities – among which are its public
status, public service constraints, or the independence of health professionals.
182
Sponsored by local and national supervisory authorities, consultants’ intervention is

often resented by hospitals’ staff: their high cost is weighed against the pointlessness
of their recommendations to achieve ‘performance leaps’.
If one now turns to the field of local authorities, a similar phenomenon is at
stake. State services and planning agencies had a quasi-monopoly on expertise
until the 1980s but decentralisation laws, the development of internal expertise
by local authorities, the investment of large consultancy firms into local public
management, pressures from the European Union, and the growing number of
public–private partnerships (PPPs) changed the situation (Abiker, 1996). Now
consultants convey a cross-sector and ‘deterritorialised’ expertise, focused on a
unique conception of ‘performance’ (Bérard, 2011) regardless of the distinctive
social, historical, economic or political features of a particular area.
In the same perspective, consultants translate issues into their own categories
of understanding and transcribe situations into their corporate language in order
to reduce uncertainties or dilute any specificity into standardised categories and
handy one-size-fits-all solutions. In other words, they format the thinkable.
Being confined to a methodological dimension, consultants seemingly comply
with their principals’ autonomy. The former apparently fades behind the latter.
Consultants tend indeed to present the methodological skills and tools they
sell to executives as politically neutral ones, their neutrality justifying their
high level of formalisation and standardisation, as well as their transferability
from one sector to another. But as with any other policy tools, consultants’
instruments are obviously not neutral (Chiappello and Gilbert, 2013). First,
their transferability only goes one way: from the private sector, adorned with all
the virtues, to the public sector, which organisation and working methods are
deemed as disqualified. Therefore, and second, these tools implicitly convey the
dominant rationality of their place of manufacturing, namely industrial rationality.
They produce effects of truth and interpretation of the world. Specifically, they
convey a managerialist and productionist philosophy. In the hospital sector
for example, under the guise of promoting ‘quality’ in emergency services,
consultants sent by the National Agency for Support to Performance (ANAP)
generated productivity indicators – for waiting time in casualty departments, for
example (Belorgey, 2010). Presenting quantitative indicators as qualitative ones
partly countered medical oppositions: no professional could openly oppose the
official objective of improving the quality of the services he or she delivered. In
general, benchmarking indicators produced by consultants encapsulate arbitrary
conceptions of organisations and public activities (Bruno and Didier, 2013). In
the case of PFI, the high technicality of the financial instruments endows the
consultants who master them with control over policymakers. Along the way, the
complexity and technicality involved played a part in naturalising the requirements
of lenders and investors. Technical tools transform public facilities into financial
assets and consultants create ‘good [public] clients’, in line with financial markets’
standards (Deffontaine, 2012).
183
Opposing consultants’ intervention is particularly difficult as they are mobilised

and supported by administrative and political elites, as they develop esoteric
language and instruments, and find objective allies among dominated subjects in
workplaces undergoing ‘modernisation’. Admittedly, consultants are ridiculed
because they are expensive and because they are ignorant of the sectors which
they reform. This satire however has not stopped the normalisation and everyday
usage of their services. Consultants’ intervention drove labour unions to shift
their approach from contentious to managerial. Thus, some labour unions have
developed a ‘counter-expertise’ through specialised firms. For example, France
Telecom Orange, which used to be a public utility (Postes et Télécommunications
[PTT]) before its privatisation in 1996, since had to face an increasing number
of suicides from its employees subjected to continuous restructuring plans. These
suicides led labour unions to resort to consulting firms which shared their views
in order to develop an expertise in ‘psychosocial risks’. Union counter-expertise
does empower employees to challenge consultants in charge of restructuring plans.
However, resorting to this type of collective action strategy is likely to further
isolate the leadership from the grassroots and weaken union solidarity. Thus,
struggles about working conditions increasingly oppose experts of one kind or
another, while employees barely have any influence over them. Caught by the
competition at work in the consulting business, some consulting firms created by
large groups of affiliated trade unions are even tempted to play the competitive
game with incumbent consulting firms (Henry, 2012).
In the end, consultants are key players of the ‘mimetic institutional isomorphism
process’ (Di Maggio and Powell, 1983). By crossing organisational boundaries
and circulating highly formalised tools and standardised practices, they contribute
to the structuring of ‘organisational fields’. They participate both in the de-
differentiation of the state and the loss of organisational diversity within the
state. They are key operators of the ‘neoliberal bureaucratisation’ of the world
(Hibou, 2012) and they are involved in the dynamic of ‘growing integration’ – as
understood in the theory and history of firms (Chandler, 1977) – of ‘professional
bureaucracies’ (Benamouzig and Pierru, 2012).
Meanwhile, the spread of the same standardised methods such as lean
management ignores the specific nature of certain social practices. This is
particularly true in all areas where ‘prudential practice professions’, such as
medicine, teaching or social work, operate (Champy, 2011). Professionals
in such fields must enjoy a certain autonomy of thinking because they work
on singular, complex and uncertain cases. The dissemination of standardised
working methods which closely monitor professional autonomy is a hindrance
to what Aristotle called ‘phronesis’ (practical wisdom). In doing so, consultants
contribute to the degradation of the quality of professional services while claiming
to do the opposite. It is no coincidence that the rise of large consulting firms
in public policy was contemporary to the convergence of various professional
mobilisations in the ‘Appel des appels’ (Gori et al, 2009). In the case of hospitals,
the RGPP caused an unprecedented mobilisation of the medical elite who as a
184
result were facing a closed decision-making process while this professional group
had always had direct access to policymakers (Pierru, 2013). Medical ‘bosses’ in
French public hospitals denounced the takeover by hospital managers, and the
growing intrusion of highly paid consultants while doctors and nurses’ jobs were
being cut. More generally, professionals denounce the proliferation of ‘dispositifs’
which frame professional practices from a distance and harm the realisation of
their work. Consultants nowadays symbolise the dedication of this ‘disembodied
management’ (Dujarier, 2015). The contemporary French situation in regards to
consultants’ involvement in public policies seems to contradict both the thesis of
a widespread ‘consultocracy’ and that of a national resistance to public policies’
penetration by private sector’s ‘input’ – or the so-called ‘French exception’ (Chafer
and Godin, 2005). Nevertheless, the recent changes bear witness to the increasing
deployment of market-based solutions provided by consultants to the French
public sector. Such a trend raises broader issues that pose a significant challenge
for those who aim at analysing these evolutions. It has become indeed increasingly
important to assess whether private consultants’ goals, interests and values can fit
with the non-profit mission of the public sector and its critical requirements in
terms of equality, social justice and fairness. It also raises the larger question of
democratic control over public policies’ orientation and whether decision makers
and elected representatives can genuinely continue steering while increasingly
letting consultants take over the rowing.
References
Abiker, D. (1996) Les consultants dans les collectivités locales, Paris: Librairie Générale
de Droit et de Jurisprudence (LGDJ).
Belorgey, N. (2010) L’hôpital sous pression: Enquête sur le nouveau management public,
Benamouzig, D., Pierru, F. (2012) ‘Professionals and the “System”: The
institutional integration of the medical world’, in P. Bezes, D. Demazière, T.
Le Bianic, C. Paradeise, R. Normand, D. Benamouzig, F. Pierru, J. Evetts (eds)
‘Debate: New public management and professionals in the public sector. What
new patterns beyond opposition?’, Sociologie du travail 54, English Supplement 1.
Bérard, Y. (2011) ‘Consultant’, in A. Cole, S. Guigner, R. Pasquier (eds)
Dictionnaire des politiques territoriales, pp. 86–90, Paris: Presses de Sciences Po.
Berrebi-Hoffmann, I., Grémion, P. (2009) ‘Elites intellectuelles et réforme
de l’État: Esquisse en trois temps d’un déplacement de l’expertise’, Cahiers
internationaux de sociologie 126(1), 39–59.
Berrebi-Hoffmann, I., Lallement, M. (2009) ‘À quoi servent les experts?’, Cahiers
internationaux de sociologie 126(1), 5–12.
Bezes, P. (2012) ‘État, experts et savoirs néo-managériaux : Les producteurs et
Bonelli, L. (2009) ‘Quand les consultants se saisissent de la Sécurité urbaine’,
Savoir/Agir 9, 17–28.
185
Boni-Le Goff, I. (2012) ‘“Trois mois pour faire la difference”: Usages et effets
des instruments d’intervention dans le conseil en management auprès du secteur
public’, Politiques et management public 29(1), 99–112.
Boussard, V., Martin, E.,Vézinat, N. (2015) ‘La permanence dans le changement.
Les usages renouvelés de la “modernisation” des entreprises publiques’, Sociétés
contemporaines 97, 5–23.
Bruno, I., Didier, E. (2013) Benchmarking: L’État sous pression statistique, Paris:
La Découverte.
Bustelo, M., Ferguson, L., Forest, M. (eds) (2016) The Politics of Feminist Knowledge
Transfer: Gender Training and Gender Expertise, Basingstoke and New York:
Palgrave Macmillan.
Chafer, T., Godin, E. (2005) The French Exception, New York and Oxford:
Berghahn Books.
Champy, F. (2011) Nouvelle théorie sociologique des professions, Paris: Presses
Chandler, A.D., Jr (1977) The Visible Hand: The Managerial Revolution in American
Business, Cambridge, MA and London: The Belknap Press of Harvard University
Press.
Chiapello, E., Gilbert, P. (2013) Sociologie des outils de gestion, Paris: La Découverte.
Davids, T., Eerdewijk van, A. (2016) ‘The smothering of feminist knowledge:
Gender mainstreaming articulated through neoliberal governmentalities’, in M.
Bustelo, L. Ferguson, M. Forest (eds) The Politics of Feminist Knowledge Transfer:
Gender Training and Gender Expertise, pp. 80–96, Basingstoke and New York:
Palgrave Macmillan.
Deffontaine, G. (2012) ‘Les consultants dans les PPP: Entre expertise au service
du client public et intermédiation pour protéger le “marché”’, Politiques et
management public 29(1): 113–33.
Di Maggio, J.P., Powell, W.W. (1983) ‘The iron cage revisited: Institutional
isomorphism and collective rationality in organizational fields’, American
Sociological Review 48(2), 147–60.
Dujarier, A.-M. (2015) Le management désincarné: Enquête sur les nouveaux cadres
du travail, Paris: La Découverte.
Gantenbein, M. (1993) ‘Un métier de rêve: Regards dans les coulisses du métier
de consultant’, Actes de la recherche en sciences sociales 98, 69–80.
Gervais, J. (2007) ‘Former des hauts fonctionnaires techniques comme des
managers de l’action publique. L’“identité managériale”, le corps des Ponts et
Chaussées et son rapport à l’État’, Politix 20(79), 101–23.
Gervais, J. (2012) ‘Les sommets très privés de l’État: Le “Club des acteurs de la
modernisation” et l’hybridation des élites’, Actes de la recherche en sciences sociales
194, 4–21.
Gori, R., Cassin, B., Laval, C. (2009) L’appel des appels. Pour une insurrection des
consciences, Paris: Mille et une nuits.
Henry, O. (1992) ‘Entre savoir et pouvoir: Les professionnels de l’expertise et du
conseil’, Actes de la recherche en sciences sociales 95, 37–54.
186
Henry, O. (2012) ‘Les syndicats et l’expertise en risques psychosociaux: Note

de recherche sur les années noires du management à France Télécom Orange’,
Actes de la recherche en sciences sociales 194, 52–61.
Henry, O., Pierru, F. (2012) ‘Les consultants et la réforme des services publics’,
Hibou, B. (2012) La bureaucratisation du monde à l’ère néolibérale, Paris: La
Découverte.
Hoard, S. (2015) Gender expertise and public policy: Towards a theory of policy success,
Basingstoke and New York, Palgrave Macmillan.
Hood, C., Jackson, M. (1991) Administrative argument, Dartmouth: Aldershot.
Linossier, R. (2012) ‘Le conseil en stratégies et projets urbains: Un marché
atypique’, Politiques et management public 29(1), 57–78.
Magnin, B. (2010) ‘La réussite du thème de la crise de l’État ou l’impérialisme
de la compétence’, in L. Bonelli, W. Pelletier (eds) L’État démantelé, pp. 54–64,
Maillard, J. de (2009) ‘Réformes des polices dans les États occidentaux: Une
perspective comparée’, Revue française de science politique 59(6), 1197–230.
Pierru, F. (2012) ‘Le mandarin, le gestionnaire et le consultant: Le tournant
néolibéral de la politique hospitalière’, Actes de la recherche en sciences sociales
194, 32–51.
Pierru, F. (2013) ‘Les mandarins à l’assaut de l’usine à soins: Bureaucratisation
néolibérale de l’hôpital et mobilisation de l’élite hospitalo-universitaire en
France’, in B. Hibou (ed) La bureaucratisation néolibérale, pp. 203–30, Paris: La
Découverte.
Poupeau, F-M., Guéranger, D., Cadiou, S. (2012) ‘Les consultants font-ils (de)
la politique ?’, Politiques et management public 29(1), 9–19.
Rottenberg, C. (2014) ‘The rise of neoliberal feminism’, Cultural studies 28(3),
418–37.
Saint-Martin, D. (2000a) Building the new managerialist state: Consultants and the
politics of public sector reform in comparative perspective, Oxford and New York:
Saint-Martin, D. (2000b) ‘Enarchie contre ‘consultocratie’: Les consultants et
la réforme administrative en France depuis les années 80’, Entreprise et Histoire
25, 82–92.
Sorignet, P.-E. (2012) ‘“On n’y comprend rien”: Des salariés européens face à
l’action des cabinets de conseil dans la réforme de l’audiovisuel public’, Actes
Thine, S. (2006) ‘Pierre Bourdieu: Éléments d’une analyse du champ du conseil’,
Revue française de gestion 165, 35–43.
Thine, S. (2012) ‘Du militant au manager’, Actes de la recherche en sciences sociales
193, 74–79.
Vezinat, N. (2015) ‘Stratégies, formes et facteurs de la modernisation postale:
Regard sur la financiarisation des PTT’, Sociétés contemporaines 98, 25–48.
187
Part Four
Parties, interest groups,
research institutes
and think tanks
189
ELEVEN
The field of state expertise

Mathieu Hauchecorne and Etienne Penissat
Unlike other countries where public policy expertise comes primarily from
academia or think tanks (Desrosière, 1999; Medvetz, 2012), governmental
administration in France has historically been a privileged site for the production
of knowledge pertaining to the state and its interventions.1 This state expertise
consists of myriad recommendations from experts and research institutes affiliated
with various ministries and the Prime Minister’s office. Taken together, they
are the primary site for the production of statistical knowledge of the French
economy and society, and the main source of administrative reports intended
to guide and evaluate state intervention. Most knowledge of the state, which
could be considered the French equivalent of American ‘policy analysis’, is
produced in these institutions. Today this field of state expertise bears the legacy
of three reconfiguration processes. The first consists of the establishment of
indicative planning after the Second World War, followed by the liberal turn
economic policy began to take in the mid-1970s. The second is associated with
the transformation of the top tier of the French civil service, within which the
great corps of engineers (graduates of two exclusive state technical schools,
commonly known as ‘X-Mines’ and ‘X-Ponts’) first assumed responsibility for
state expertise, only to face competition subsequently from the corps of statisticians
(administrators of the French statistics institute (Institut National de la Statistique
et des Études Économiques [INSEE]) and the administrative corps (financial and
social affairs inspectors). The field of French state expertise also shows traces of
the institutionalisation process experienced by the social sciences, which initially
took place on the fringe of French universities.
State reports and the institutional logics underlying their being requested, their
creation, and their reception cannot be understood without considering the
structure of the field of state expertise in France and how it has changed since
1945. This is obviously true for reports coming from councils, commissions and
research services affiliated with various ministries and the Prime Minister’s office,
but it is also true of reports produced by the ad hoc commissions to which the
French President, Prime Minister and Ministers may turn from time to time,
since their work tends to rely heavily on the work of these longer-lasting bodies.
Thus, although the founding and transformation of various bodies and councils
within the state owe much to political will, the field of state expertise that they
shape nonetheless enjoys relative autonomy, which explains how the ‘requests’ for
such reports can often be partly produced by these bodies themselves. Although
it is structured around specific issues and oppositions, the field of state expertise
191
must be understood in relation to three other spaces: the political field, of course,
to which the reports produced by these bodies and councils are supposed to be
addressed; the intellectual field, of which the field of state expertise is a segment;
and the economic field and the range of interest groups (employers or workers’
unions, non-profit groups, and so on) likely to be consulted during the preparation
of these reports.
As we will see in the first section, the field of state expertise was formed
in the first two decades following the Second World War, largely in response
to the needs of French indicative planning. Initially strongly centred on the
General Planning Commission (Commissariat Général du Plan [CGP]), it gradually
developed a polyarchical structure with the development of ministerial research
services and the emergence of challenges to planning (second section), which
favoured the reorientation of expertise toward the measurement and analysis of
state intervention through public policy evaluation. The last two decades were
marked by the revocation of the Plan and the creation of new councils, lighter
organisations over which political authorities could exercise greater control (third
section). A means of bypassing the Plan’s modernisation commissions in the
post-war years, the government-appointed ad hoc commissions and the reports
they produced took on new meaning in this changed context (fourth section).
The building of a field of state expertise (1945–65)

The production of reports within the state is an old practice whose first examples
can be traced back to the nineteenth century, and the roots of many research
services at public authorities’ disposal today lie in the political reconfigurations
of the 1930s and the war. But a genuine field of state expertise did not take
shape until the decades following the Second World War, changing scale and
becoming institutionalised with the creation of several research services. These
transformations were related to the development of indicative planning and
its need for more standardised knowledge of French society, its economy and
the national territory. The institutional scheme comprised of the Plan and the
numerous research services gravitating around it is thus probably intrinsically
tied to the expanding reach of state economic intervention. It also reflects the
emergence of new patterns of legitimation for public authorities, as Delphine
Dulong has demonstrated. Faced with the legicentric model making parliament
and the law the cornerstone of state intervention, post-war ‘modernisers’ backed
an initiative for rational policy, rooted in economic expertise and forecasting,
to be implemented by re-enforced administrative and executive branch powers
(Dulong, 1997).
Created in 1946, the CGP was originally charged with three functions:
improving knowledge of the French economy and society, defining the
medium- and long-term objectives of economic policy, and proposing measures
to reach these objectives (Bauchet, 1970). Initially backed by a small elite from
the state’s high-ranking technical corps, the newly founded National School of
192
Administration (École Nationale d’Administration), and (to a lesser extent) industry

and unions (Dulong, 1997; Rousso and Margairaz, 1987), the CGP was formed
around the commissioner, a team of several dozen civil servants, and modernisation
commissions whose reports and recommendations were supposed to contribute
to the definition of four- to six-year Plans. Commissions were composed of one
third representatives of the administration, one third representatives of employers
and workers, and one third experts (Bauchet, 1970).
The output of these commissions was sometimes prepared by less formal
working groups, usually formed within the Plan’s administration (Bauchet,
1970). They also relied on analyses produced by a whole string of institutions
external to the Plan but intervening in the elaboration and/or implementation
of successive Plans, especially INSEE and the Economic and Financial Studies
Service (Service des Etudes Economiques et Financières [SEEF]), created in the Ministry
of Finance in 1952 (Bauchet, 1970, 60; Fourquet, 1980). In addition to a shared
worldview or even ‘rationalist ideology’ (Jobert, 1981), the connections between
these organisations are notably manifest in the rank held by actors involved in
planning and their circulation between these institutions (Dulong, 1997). A kind
of national account system took shape through such staff exchanges between the
CGP, SEEF and INSEE, making it possible to represent the French economy
and make it a subject of state intervention (Fourquet, 1980).
The CGP was additionally driven to create and rely on a proliferation of newly
created research bodies whose work was supposed to contribute to preparation
of the Plan (Bauchet, 1986, 25). The Centre for Consumption Research and
Documentation (Centre de Recherche et de Documentation sur la Consommation
[CREDOC]) was founded in 1954 to study household consumption (Benamouzig
2005). The scenarios submitted to modernisation commissions could be developed
using macroeconomic models developed by the Centre for Economic and
Mathematical Forecasting Studies Applied to Planning (Centre d’Etudes Prospectives
Economiques et Mathématiques Appliquées à la Planification [CEPREMAP]), which
quickly became an interface between state expertise and fundamental research
(Fourquet, 1980, 284–9). Studies devoted to regional development were confided
to the Delegation for Land-Use Planning and Regional Intervention (Délégation
à l’Aménagement du Territoire et à l’Action Régional [DATAR]), created in 1963,
which helped develop human geography (Claval, 1998). Since 1966 the Centre
for the Study of Revenues and Costs (Centre d’Etude des Revenus et des Coûts
[CERC]) produces reports on developments in prices, incomes and inequalities.
The Centre for Forecasting and International Information Studies (Centre d’Etudes
Prospectives et d’Informations Internationales [CEPII]) was added to the ranks of
these organisations in 1978, to address the international economy. Within the
Plan and these organisations emerged the figure of the engineer-economist, most
often an X-Mines or X-Ponts alumnus, who promoted a formalised economic
approach at a time when the economic sciences being taught in law programmes
still had a very literary character (Fourcade, 2009). Some new areas of analysis
emerged or developed, including public economics, which applies the tools of
193
microeconomics to the study of the role of the state, and health economics, which
developed in various forms within CREDOC and the Ministry of the Economy
and Finance (Benamouzig, 2005).
Although the Plan was a linchpin of state expertise, it should be emphasised
that many research services and public bodies of experts began to appear in
governmental ministries in the post-war period. This is notably the case of
the National Institute for Demographic Studies (Institut National d’Etudes
Démographiques [INED]), founded in 1945, which helped establish family policy
based on statistics and demography (Rosental, 2003) and even more so for INSEE,
which transformed pre-war statistical production, criticised for being extremely
deficient and incomplete, into a solid public statistics system producing social and
economic data supporting planning policy. The originality of these institutes must
be signalled from the outset: while most European statistical institutes are dedicated
exclusively to the production of data to be analysed by other organisations and
academics in their own studies, in France statistics, research and analysis are
integrated (Desrosières, 2000). So INSEE administrators, a state corps mainly
composed of graduates from the prestigious École Polytechnique (an elite technical
school), present themselves as ‘magistrates of numbers’ as much as economists,
producing both macro-economic predictive models and in-depth studies of the
transformations in French society. This explains why these institutes acquired
immediate visibility and significant stature in the field of expertise in France.
Additionally, INSEE’s position is even more important because its administrative
corps circulates through a variety of ministries.
Alongside these major institutes, some ministries have their own statistical
service as well as a research service or fully-fledged institute, and sometimes
even their own educational institution. The statistical division of the Ministry of
Labour was confided to a young INSEE administrator in 1946; the Centre for
Social and International Relations Research (Centre de Recherches Sociales et des
Relations Internationales) was given to the Marxist economist Charles Bettelheim
(Denord and Zunigo, 2005); and the Social Sciences and Labour Institute (Institut
des Sciences Sociales et du Travail), whose vocation was to train unionists, was created
with the joint oversight of the University of Paris in 1951 (Tanguy, 2008).
This last example thus indicates that there was a potential place for the social
sciences among the expertise acquired by Ministries, and sociology in particular.
This opening nonetheless remained marginal, and state expertise was still primarily
concentrated around the Plan, SEEF and INSEE.
From the ‘crisis of the Plan’ to the neoliberal turn (1965–95)

The field of state expertise that emerged in the post-war years was deeply
transformed by reconfigurations that began to affect French planning in the mid-
to late 1960s. Several factors contributed to redefining the context structuring
modernisation commissions and imposed the ‘crisis of the Plan’ or ‘unplanning’
(déplanification) theme on French economic policy producers and analysts. This
194
new realignment is first of all inseparable from the marked reconfiguration of

senior civil service at the Plan level, as modernisation commissions included
increasing numbers of ‘enarques’ (graduates of the highly selective National
Administration School [École Nationale d’Administration]), finance inspectors and
INSEE administrators, to the detriment of X-Mines and X-Ponts alumni and
representatives from technically oriented ministries (Rousso and Margairaz, 1987).
It also came from transformations in the political field, as De Gaulle’s departure
from the French presidency and Valéry Giscard d’Estaing’s arrival in power in
1974 favoured a liberal shift in economic policy (Fourquet, 1980, 291–5). Last,
it is connected to changes specific to the economic field: if there was a tendency
toward less representation of industrial, and even labour union, interests in Plan
commissions after 1965, banking sector involvement increased, while the private
sector increasingly took the Treasury’s place in funding public investments
(Andrieu, 1987). These reconfigurations are moreover not without consequence
on the Plan’s orientations: the ‘revenue policy’, of Keynesian inspiration, was
challenged following the departure of Pierre Massé in 1965 (Fourquet, 1980), and
the Finance Commission supported the introduction of Milton Friedman’s work
and monetarist ideas into state expertise (Andrieu, 1987, 169). The state’s shift
to neoliberal approaches led experts to focus more on the state, its effectiveness
and its functioning, and make it an object of knowledge (Bezes, 2002; 2009).
It all adds up to make the state, subject to criticism and a reform process, into a
political subject.
The Plan’s decreased legitimacy in steering the economy seemed to initially
contribute to strengthening its research-related functions. The central importance
accrued to its study functions was especially manifest in the June 1969 creation
of the Centre for the Coordination and Orientation of Economic and Social
Development Research (Centre de Coordination et d’Orientation des Recherches sur
le Développement Economique et Social [CORDES]). The Centre differed in two
ways from other research bodies at the time. Instead of producing research as
such, it funded studies by requesting research grant proposals on particular themes.
Organised around an ‘epistemic community’ of high-ranking civil servants and
researchers, it connected political demand for applied research with the interests
of researchers in French universities undergoing massification (Bezes and de
Montricher, 2005). The Plan’s repositioning encouraged sectorial ministries to
develop their own capacities in expertise. Their research services experienced a
period of growth in the late 1960s and 1970s. From a few dozen agents in the
1960s, the research and statistics services of the Ministries of Labour and National
Education rose to nearly 200 agents each in the late 1980s (Penissat, 2009; Pons,
2008). This context lead to the permanent institution of statistics and economics
at every level of the state in the ‘modernising’ Fifth Republic (Dulong, 1997)
and the ‘colonisation’ of various ministerial statistics and research services by
INSEE administrators. But at the Plan’s incentive, ministries also opened up to
the social sciences to an unprecedented extent. Research funding thus played
a key role in shaping a vast field of empirical research in France, especially
195
in sociology (Pollak, 1976; Amiot, 1986). These ministerial services brought

together INSEE statisticians, a full range of staffers from the ministry in question,
and a great many people working on contract recruited from universities and
research centres. Although these different kinds of agents rarely worked together
at the beginning, their diverse skillsets gradually started to converge. This was
notable in the production of statistical surveys: starting in the 1970s, INSEE’s
great cyclical surveys and administrative data sources would be supplemented by
thematic surveys using specialised questionnaires and innovative methods (panels,
following cohorts and so on). The findings of such research still took report form,
but were also promoted in formats similar to those used in academia, especially
journals. The social sciences thus underwent a movement of institutionalisation
and independence (Bezes et al, 2005). Researchers and analysts in these services
were in a split position, enlisted in backing the development and support of
sectorial policies while also contributing to the dissemination of social critique
(Pollak, 1976). This movement is found in relation to the Plan and its associated
organisations, as the studies they funded and developed helped redefine planning.
If the Plan’s early years were marked by the development of national account
systems, econometric advances in the late 1960s and early 1970s made it possible
to model a variety of economic policy scenarios, as exemplified by the Physico-
financial model (FIFI) (Angeletti, 2011). Similarly, the production of indicators
was a response to the growing importance accorded to the evaluation of public
policy (Jobert, 1981; Spenlehauer, 1998).
This movement mainly concerned the Plan and the so-called ‘social’ ministries
(Health, Education, Labour and Employment), whose need for this research was
even greater because they were trying to use science to build their legitimacy to
defend themselves from the Ministry of Finance, which increasingly controlled
their interventions. In fact, the Ministry of Finance created the Forecasting
Division for that very purpose in 1965, replacing the SEEF (Terray, 2003).
While continuing to develop expertise in economic forecasting, the division also
staffed itself with project managers that developed cost–benefit analyses of the
interventions of other ministries through what was known as the Rationalisation of
Budgetary Choice (RBC), imported from the United States (Spenlehauer, 1998).
Indeed, the top civil servants of the so-called ‘transversal’ ministries (Finances,
but also Interior and Civil Service) tried to concentrate and get control over
the expansion of areas of state intervention, especially in educational, health and
social policies. They did this in part by developing knowledge sets aiming to
audit management and evaluate the effectiveness of their public policies, with
particular reliance on microeconomics. Although the RBC failed, in no small
part due to the resistance of the corps of the state, this ‘state self-concern’ would
simply shift to public policy evaluation in the 1980s (Bezes, 2002).
Recourse to social science or economic expertise has been more marginal in
‘sovereign’ ministries (Justice, Interior, Foreign Affairs, Defence). The research
services in these ministries are usually relatively small, as is the case for the Centre
for Social Science Studies of Defence (Centre d’Etudes en Sciences Sociales de la
196
Défense [C2SD]), a research body contractually affiliated with the Ministry of

Defence (Jankowksi and Vennesson, 2005), or the Analysis and Forecasting Centre
(Centre d’Analyse et de Prévision [CAP]), a small research unit at the Ministry of
Foreign Affairs (Kessler, 2005). These ministries make flexible use of expertise
from small services or a handful of institutions of various statuses in the Ministry’s
remit, as seen with the Justice Ministry’s statistical service, the Law and Justice
Mission (Mission Droit et Justice), that funds studies, and the Centre for Sociological
Research on Law and Penal Institutions (Centre de Recherche Sociologique sur le Droit
et les Institutions Pénales [CESDIP]), which maintains a privileged relationship with
the Ministry despite having become independent in the 1980s. Their status is
also more fragile and less stable, depending on political circumstances – the left
encouraging their creation, the right neglecting them – or even specific political
figures’ appetite for studies. Alain Joxe provides a prime example: a Minister from
the Socialist Party, he promoted the creation of the National Institute of Higher
Internal Security Studies (l’Institut Nationale des Hautes Etudes de la Sécurité Intérieure
[IHESI]) at the Ministry of the Interior in 1989, and then the 1992 founding
of the Strategic Affairs Delegation (Délégation aux Affaires Stratégiques [DAS]), a
sort of think tank charged with ordering studies for the Ministry of Defence
(Frison-Roche, 2005). IHESI illustrates a configuration where political changes
considerably affect the position of both the content and institution of expertise:
founded to defuse security debates by introducing an expert discourse and more
specific consideration of the history and sociology of policing, it was challenged
when the right came to power in 1993 and Charles Pasqua became Minister
of the Interior, because it was considered to be too critical of the police corps.
The institute was renewed when the left rose to power in 1998, only to once
again be jeopardised by Nicolas Sarkozy’s arrival as head of the Ministry in 2002,
especially due to his preference for using private expertise firms (Ocqueteau and
Monjardet, 2005). With the exception of the Ministry of Justice, acceptance of
social science research has often proved to be tenuous in these ministries, where
civil servants often consider themselves experts in their domains and more readily
work piecemeal with private firms (the case for the Ministries of Defence and
Foreign Affairs), or where they more explicitly reject the social sciences, as with
the police and the prefectoral corps in the Ministry of the Interior.
If these organisational reconfigurations of state expertise services favour a greater
plurality of disciplines and knowledge, the neoliberal turn of the 1980s sped up
the marginalisation of the Plan in the field of state expertise that had begun in the
mid-1970s (Jobert and Théret, 1994). CORDES was eliminated in 1979 and the
Plan commission structure was modified in 1982 by the creation of the National
Planning Commission (Commission Nationale de la Planification), which notably
made room for representatives from the non-profit sector (Bauchet, 1986, 22).
The 1980s were thus an ambivalent period for the structure of state expertise.
The left’s 1981 arrival in power allowed contractual agents hired in the 1960s
to 1970s to accede to permanent positions, which stabilised ministerial research
service staff. Likewise, the position of these services is significantly strengthened
197
in both ministries and in the field of state expertise. The Ministry of Finance’s
increasing intervention in the policies of sectorial ministries through its Forecasting
Division (Jobert and Théret, 1994) indeed pressed these other ministries to
equip themselves with their own expert capacities. In some cases, this leads to
the creation of full-fledged ministerial divisions: the Evaluation, Forecasting,
and Performance Division (currently called the Direction de l’Evaluation, de la
Prospective, et de la Performance [DEP]) at the Ministry of National Education
in 1987, Research, Studies, and Statistics Division (Direction de l’Animation de
la Recherché, des Etudes, et des Statistiques [DARES]) at the Ministry of Labour
in 1992, and, later on, Research, Studies, Evaluation, and Statistics Division
(Direction de la Recherche, des Etudes, de l’Evaluation, et des Statistiques [DREES]) at
the Ministry of Health in 1998 (Serré, 2001). Finally, the increasingly demanding
standards of public policy evaluation seemed, at least at the outset, to offer a new
space in which these services could invest. Staff were able to benefit from it by
obtaining new means for research and funding studies while retaining the ability to
critique governmental intervention. Moreover, CGP leadership in the 1980s and
early 1990s tried to re-focus the Plan on public policy evaluation, experiencing
a boom at the time. This strategy ran up against the opposition of a variety of
administrative and political actors (especially in the senior civil service) worried
about a potential renewal of influence making it possible to coordinate public
policy through evaluation (Spenlehauer, 1998).
A rising politicisation of state expertise? (1995 to the present)

Preparation of the eleventh Plan was abandoned in 1993, leading to the CGP’s
disappearance a few years later. This put the field of expertise through a new
phase, marked by the multiplication and the accumulation of public expertise
organisations.
After 1995, several reports and position statements were applied along these
lines to reconceive the configuration of research services affiliated with the Prime
Minister, by, for example, charging the Plan with orchestrating public debate by
framing enquiry and disseminating arguments and data likely to contribute to
deliberation, using the expertise of its affiliated organisations. This reconfiguration
caused some organisations to disappear, others to be transformed, and new ones to
be created, such as the Economic Analysis Council (Conseil d’Analyse Economique)
in 1997, the Pension Orientation Council (Conseil d’Orientation des Retraites)
in 2000, and the Employment Orientation Council (Conseil d’Orientation pour
l’Emploi) in 2005.
These new structures differed on many points from the former modernisation
commissions and, to a lesser extent, the research services associated with the Plan.
Unlike modernisation commissions, they left barely any place for representatives
of economic or union interests, appearing to be more turned toward expertise
in consequence. This new orientation proceeded to make more room for actors
from outside the administration. This was especially true in the Economic Analysis
198
Council, mainly composed of university-based economists: its creation came to

challenge the monopoly over economic expertise previously held by members of
the senior civil service (Clairat, 2013). In addition, these various ‘councils’ were
lighter structures than older Plan organisations, primarily charged with using grant
offers to spur research and studies in other institutions such as INSEE, INED,
universities, or the National Centre for Scientific Research (Centre National de la
Recherche Scientifique [CNRS]). In these regards, the restructuring of state expertise
bore the imprint of economics and the social sciences’ institutionalisation in
the university system and redefined power relations between high-ranking civil
servants and academic researchers. It also bore the mark of new public management
(Bezes, 2005), as these light structures seem more subordinate to public demand.
The 1993 elimination of the CERC and the morphology of the organisations
that succeeded it are typical of these developments. Founded in 1966, CERC
predominantly studied the evolution of income and capital inequalities, and its
reports from the early 1990s sparked debate upon several occasions. Markedly
situated to the left, it was eliminated in 1993 by the conservative Prime Minister
Edouard Balladur, who replaced it with a new organisation, the Higher Council
for Employment, Revenues, and Expenses (Conseil Supérieur de l’Emploi, des
Revenus et des Coûts [CSERC]). This flexible structure was organised around
a smaller team of rapporteurs who were generally transferred to CSERC from
elsewhere in the administration for a few years, and its main function was
producing summarising reports and encouraging the pursuit of new themes in
French public research and statistics by offering research grants on specific topics.
In 2006, the CGP was eliminated and the Strategic Analysis Centre (Centre
d’Analyse Stratégique) took its place, to be renamed Strategy France (France Stratégie)
in 2013. This new institution was given the mission of public policy evaluation,
forecasting, consulting and advising.2 To this end, it coordinated the activities
of the councils that had been founded since 1997 in affiliation with the Prime
Minister’s office. Expertise production in Strategy France consists mainly of
offering summaries and short reports on themes and issues of current relevance
to political leaders, and they are also concerned with getting this expertise into
the media and public debate.
Due to these morphological changes, state expertise seems to lose some of its
independence from political authorities. This politicisation of expertise illustrates
changes that typify this period beginning in the 1990s. Indeed, expertise was
established in the 1980s – and considered essential in the 1990s – as a veritable
tool for political communication and politicising public policy. State expertise
organisations began to publish their studies and data in many formats (aimed at
audiences from the general public to the most academic); the conference format
became increasingly widespread; ministerial cabinets repeatedly used ‘numbers’
as tools for intervening in the public debate (Tissot, 2004; Penissat, 2012). The
price of expertise’s greater visibility was a tightening of political control over the
production of studies and data that was manifest in many ways. The first was the
creation of incentivised research funding mechanisms (calls for grant proposals)
199
on topics chosen according to political concerns. Likewise, the appearance of

expert divisions in ministries mainly occurred through the creation of small
research cells to meet political demand, especially for economic and budgetary
forecasting. In some cases, these organisations were exclusively seen as resource
agencies for responding to political leaders’ requests for solutions, as is the case
for the National Institute for Public Health Surveillance (Institut National de Veille
Sanitaire) (Buton, 2006).
This re-enforced political constraint did not occur without resistance from
the organisations’ staff, and it led to many tensions that sometimes made it into
the media. Debates over the 2007 unemployment figures, recurrent disputes
over delinquency statistics, and conflict over the non-publication of studies and
reports produced by the Ministry of National Education’s expertise service all
illustrate the rising tension over the dissemination of statistics and studies (Penissat,
2009). Likewise for the controversy over the institutionalisation of criminology
in universities, which was jointly driven by political actors (Nicolas Sarkozy),
private-sector actors, and expert organisations affiliated with the Ministry of
the Interior (such as the National Institute for Higher Security Studies [Institut
National des Hautes Etudes de la Sécurité]), proof par excellence of the role of state
expertise in the institutionalisation of academic disciplines (Mucchielli, 2010).
It should lastly be stressed that the fragmentation and multiplication of
organisations does not work counter to a certain homogenisation of the knowledge
being used. To the contrary, changes in the characteristics of the hired agents
and the transformation of power relations in the scientific field were decisive in
fostering an emergent uniformity. Indeed, in the 1980s to 1990s, permanent
INSEE staffers became established as the dominant experts, while contractual
employees were marginalised. Most of the initially hired INSEE administrators
had been trained in the social sciences, and proved to be useful in countering the
economic orthodoxy movement promoted by the Ministry of Finance. Those at
the Ministry of Labour, for instance, backed the 35-hour workweek reform that
the Ministry of Finance’s experts rejected. This employee pool changed as a result
of transformations in agents’ educational backgrounds; however, econometrics
and neoclassical economics are the order of the day at the administrator training
school, the National School for Statistics and Economic Administration (École
Nationale de la Statistique et de l’Administration Économique [ENSAE]). Likewise,
in the field of economics, heterodox economists find themselves marginalised in
favour of neoliberal economists and econometrics specialists (Lebaron, 2000). This
shift is at the root of the movement to standardise expertise around econometrics
and individualist approaches that began in the 1990s.
The reconfigurations of state expertise and the creation of new organisations also
reflect the emergence of new public issues, however. The powers of the National
Consultative Ethics Committee (Comité consultatif national d’Ethique [CCNE]) –
created in 1983 – were strengthened during the 2000s as result of growing public
concern for bio-ethics issues (Memmi, 1996). Likewise, the development of state
feminism over the past decades also had an impact on the field of state expertise
200
(Revillard, 2016), especially through the experts attached to the ministry of

Social Affairs, where women have traditionally been more numerous. The first
committees on ‘women’s condition’ was set up during the 1970s. The creation
of the Observatory on Parity between Men and Women (Observatoire de la parité
entre les femmes et les hommes [OPFH]) in 1995 and its replacement in 2013 by a
national council on equality between men and women (Haut Conseil à l’égalité
entre les hommes et les femmes) shows the gradual institutionalisation of a state
expertise on gender inequalities. The feminist critique of patriarchy also had an
impact on the categories and statistical classifications promoted by state expertise
(Fouquet, 2003; Ponthieux, 2013). For instance, the category of ‘household’ used
by INSEE has gradually been reshaped in order to account more adequately for
family structures and women’s activity (De Saint Pol et al, 2004; Amossé and De
Peretti, 2011). More generally, the development and systematisation of gender
statistics was supported by a circular signed by Prime Minister Lionel Jospin on
8 March 2000. More recently, the creation of the Observatory of Secularism
(Observatoire de la laïcité) was meant to provide the government with an expertise
on religious issues.
Ad hoc commission reports

These reconfigurations of the field of state expertise naturally modify the role of
the ad hoc commissions to which the French government may turn from time
to time, as well as the reports that they produce. Although it is impossible to
enumerate them all, such reports – which may occasionally be the work of isolated
authors3 – have made a powerful contribution to defining the issues in the French
public debate and getting various measures onto the political agenda since 1945.
Officially instituted by the Prime Minister, sometimes in association with the
President or a cabinet minister, the ad hoc commissions seem to be more flexible
than the aforementioned councils and organisations, and their work can sometimes
be even more politicised. Their assigned missions, requested by the Prime Minister’s
office, are indeed directly tied to governmental strategies, and their members
generally set out to formulate a number of practical recommendations. Their
composition is less restrained than that of the Plan’s modernisation commissions
prior to 1993, and their membership reveals a wide range of options. To give but
a few examples, the committee formed around Louis Armand and Jacques Rueff
in 1959 to consider ‘obstacles to economic expansion’ included high-level civil
servants, academic figures and representatives of labour and employers’ unions
(Rueff and Armand, 1960). Presided over by the economists Jean-Paul Fitoussi,
Amartya Sen and Joseph Stiglitz, the Commission on the Measure of Economic
and Social Performance, formed in 2008 at the initiative of French president
Nicolas Sarkozy, was composed entirely of researchers and academics.4 Formed
the following year, the Commission on the Liberation of Growth (Commission
pour la libération de la croissance) presided over by the essayist Jacques Attali (2010),
accorded a much greater place to business leaders and representatives of private
201
interests. One might follow Philippe Bezes and distinguish three models, which
each commission follows to a varying degree – the scientific commission, the
technical commission and the partisan commission. They represent so many ways
of reconciling the various imperatives commission composition must respect:
the guarantee of some kind of expertise, concern for the representativeness of its
members, and their acceptability to the relevant political authorities and agents
(Bezes, 2003). The range of their missions is left to the government’s discretion
and may include all major areas of public policy, as with the Commission on
France of the Year 2000, created in 1994 around the businessman-essayist Alain
Minc (Hauchecorne, 2009), or on a specific topic, like the Mission on the
Responsibilities and Organisation of the State, formed in 1993 under Jean Picq
(Bezes, 2003). They do the work and submit their reports in a relatively short
timeframe (from a few months to a year), or at any rate shorter than the time
needed to prepare fully-fledged Plans.
With the establishment of the Plan after the war, these flexible ad hoc
commissions are sometimes seen as a way to work around the modernisation
commissions, which were considered more democratic (Andrieu, 1987, 166).
Ad hoc committees are thus constituted where distinctly political issues intersect
with issues more specific to state expertise. For example, from April to November
1994 about 20 intellectuals and a few top-level civil servants and business leaders
were brought together in the Commission on France of the Year 2000, instituted
by Prime Minister Edouard Balladur following the proposal of the General
Commissioner of the Plan at the time, Jean-Baptiste de Foucauld. De Foucauld
conceived of the commission – a small group of experts in charge of skimming
a vast number of subjects (economic policy, education, European policy and
redistribution, to name but a few) in a short time – as a possible response to the
‘crisis of the Plan’, the Plan putting its expertise and a team of rapporteurs at
the commission’s disposal. However, while the commission is presided by Alain
Minc, one of the Prime Minister’s closest counsellors, his report also proved to be
an unofficial campaign platform for Edouard Balladur, who had not yet officially
declared his candidacy in the presidential race (Hauchecorne, 2009). If they are
a way of circumventing the ordinary rationales of state expertise, these ad hoc
commissions are nonetheless closely related to them. The resources of permanent
councils and organisations are usually put at their disposal, and they are also the
most common recruiting grounds for commission rapporteurs (Andrieu, 1987,
166). This favours a cumulative aspect in their work: the first order of business in
the Picq Commission was collecting all previous administrative reports on state
reform and making an assessment report of them (Bezes, 2003), and the 2009
mission assignment addressed to Jacques Attali for composing the Commission
on the Liberation of Growth explicitly gave him the model of the 1959 Armand–
Rueff committee.
Benefiting from unprecedented publicity, these reports sometimes enjoy wide
distribution, as happened with the ‘Computerisation of Society’ report submitted
by Simon Nora and Alain Minc, which sold 100,000 copies in France and
202
was translated into several other languages (Walliser, 1989). They are a way of
influencing public debate while giving the government an opportunity to keep
its distance, since the proposals they formulate only officially commit commission
or working group members. They nonetheless frequently lend legitimacy to
public policy by showcasing the expertise upon which they are supposedly based
and by gathering a collection of disparate measures into a coherent package, as
illustrated by the case of the Commission on the Liberation of Growth, whose
recommendations run the gamut, from the status of regulated professions to the
regulation of motor coach transportation and working on Sunday. Through this
legitimating function, these commissions can sometimes favour an orientation
change in public policy, as happened with the Barre Commission on housing
in 1975, which formalised the transition in housing policy from construction
subsidies for multi-dwelling units (known as aide à la pierre) to individual support
for tenants and home buyers (called aide à la personne) (Bourdieu and Christin,
1990).
Conclusion
The political-administrative field in France thus appears to be a privileged site
for the production of public policy knowledge and its areas of intervention. This
research function, under the purview of public authorities from 1945 to this day,
favoured the constitution of a veritable field of state expertise structured around
research services and councils of experts affiliated with various ministries and the
Prime Minister’s office. Greatly polarised by the French planning system at the
outset, today it appears to be a polycentric space where large organisations such as
INSEE and INED, which have accrued considerable scientific legitimacy and built
themselves the reputation of independence, coexist with lighter structures such
as the councils around the Prime Minister involved in Strategy France. Although
the oppositions structuring this space arise in part from political oppositions or
competition between different corps of the civil service, we have seen that they
cannot solely be reduced to those factors, and may also come from issues specific
to state expertise, ingrained in its organisational history and related to legitimate
representation of the social world, public problems and the general interest.
This field of state expertise has singular importance in the history of French
social and economic sciences, in that it was a site of their institutionalisation
between 1945 and the early 1980s. Far from being a product of academic
knowledge imported into public policy from the outset, state expertise quite
often favoured the crystallisation of new knowledge (especially in economics),
first indexed on the imperatives of public policy and only subsequently receiving
academic recognition. This relationship seems to have reversed itself in recent
years, and state expertise has stopped being a site of scientific innovation as
economics and the social sciences became institutionalised in the university and
research centres. It would also be a mistake to make these organisations and
councils’ activities out to be the simple production of knowledge about public
203
policy. As sites of state knowledge of the state, their reports are products of state
needs and internal logics. Mixing empirical description, scientific objectives and
political prescriptions, they seem to constitute the state interventions that they
are also intended to analyse. This explains why they are a privileged resource for
studies devoted to the intellectual or symbolic frameworks of state intervention.
But it is still important to remember the relative independence of this space as well
as its internal differences, which in turn leads one to question these institutions’
specific position and perspective (trying to reconcile savant and political ethos,
the demands of science, and the limitations of action) and the mediations between
these sites of production and decision-making centres.
Notes
1 This chapter was translated from French by Juliette Rogers.
2 www.cae-eco.fr/Presentation.html.
3 This is the case, for instance, of the report Education et société (Education and Society) by Jacques
Lesourne and that of Simon Nora and Alain Minc on the digital sector. For more on these two
reports, see Walliser 1989 and 1995.
4 Commission on the Measure of Economic and Social Performance (Commission pour la mesure
des performances économiques et sociales), Stiglitz et al (2009).
References
Amiot, M. (1986) Contre l’État, les sociologues: Éléments pour une histoire de la
sociologie urbaine (1900–1980), pp. 47–97, Paris: Éditions de l’EHESS.
Amossé T., De Peretti, G. (2011) ‘Hommes et femmes en ménage statistique:
Une valse à trois temps’, Travail, genre et sociétés 26, 23–46.
Andrieu, C. (1987) ‘Le plan vu de la banque, 1946–1985’, in H. Rousso (ed)
La Planification en crises: 1965–1985, pp. 157–74, Paris: Centre national de la
recherche scientifique (CNRS).
Angeletti, T. (2011) ‘Faire la réalité ou s’y faire?’, Politix 95(3), 47–72.
Attali, J. (2010) Une ambition pour dix ans, Paris: La documentation française,
XO Editions.
Bauchet, P. (1970) La Planification française, du premier au sixième Plan, Paris: Seuil.
Bauchet, P. (1986) Le Plan dans l’économie française, Paris: FNSP-Economica.
Benamouzig, D. (2005) La santé au miroir de l’économie: Une histoire de l’économie
de la santé en France, Paris: Presses universitaires de France.
Bezes, P. (2002) ‘Aux origines des politiques de réforme administrative sous la
Vème République: la construction du “souci de soi de l’Etat”’, Revue française
d’administration publique 102(2), 307–25.
Bezes, P. (2003) ‘La “mission Picq” ou la tentation de l’architecte: Les hauts
fonctionnaires dans la réforme de l’Etat’, in O. Nay, A. Smith (eds) Le
gouvernement du compromi: Courtiers et généralistes dans l’action publique, Paris:
Economica.
Bezes, P. (2005) ‘Le renouveau du contrôle des bureaucraties: L’impact du New
Public Management’, Informations Sociales 126, 26–37.
204

Bezes, P., Montricher, N. (2005) ‘Le moment CORDES (1966–1979)’, in P.
Bezes, M. Chauvière, J. Chevallier, N. de Montricher, F. Ocqueteau (eds) L’État
à l’épreuve des sciences sociales: La fonction recherche dans les administrations sous la
Vème République, pp. 37–71, Paris: La Découverte.
Bezes, P., Chauvière, M., Chevallier, J., Montricher, N., Ocqueteau, F. (2005)
L’État à l’épreuve des sciences sociales: La fonction recherche dans les administrations
sous la Vème République, Paris: La Découverte.
Bourdieu, P., Christin, R. (1990) ‘La construction du marché: Le champ
administratif et la production de la “politique du lodgement”’, Actes de la recherche
en sciences sociales 81(1), 65–85.
Buton, F. (2006) ‘De l’expertise scientifique à l’intelligence épidémiologique:
L’activité de veille sanitaire’, Genèses 65, 71–91.
Clairat, J. (2013) ‘Une mutation de l’expertise économique d’État en France?
Les nouveaux “conseillers du Prince” du Conseil d’analyse économique (1997–
2008)’, in M. Kaluszynski, R. Payre (eds) Savoirs de gouvernement: Circulation(s),
traduction(s), réception(s), pp. 213–45, Paris: Economica.
Claval, P. (1998) Histoire de la Géographie française de 1870 à nos jours, Paris: Nathan.
Denord, F., Zunigo, X. (2005) ‘Révolutionnairement vôtre’, Actes de la Recherche
en Sciences Sociales 158, 8–29.
De Saint Pol, T., Deney, A., Monso, O. (2004) ‘Ménage et chef de ménage: Deux
notions bien ancrées’, Travail, genre et sociétés 11(1), 63–92.
Desrosières, A. (1999) ‘La commission et l’équation: Une comparaison des Plans
français et néerlandais entre 1945 et 1980’, Genèses 34(1), 28–52.
Desrosières, A. (2000) La politique des grands nombres: Histoire de la raison statistique,
Dulong, D. (1997) Moderniser la politique: Aux origines de la Ve République, Paris:
L’Harmattan.
Fouquet, A. (2003) ‘La statistique saisie par le genre’, in J. Laufer, C. Marry,
M. Maruani (eds) Le travail du genre: Les sciences sociales à l’épreuve du travail des
différences de sexe, pp. 280–98, Paris: La Découverte.
Fourcade, M. (2009) Economists and Societies: Discipline and Profession in the United
States, Britain, and France, 1890s to 1990s, Princeton, NJ: Princeton University
Press.
et du Plan, Paris: Recherches.
Frison-Roche, F. (2005) ‘Des savoirs experts au service de la Défense’, in P. Bezes,
M. Chauvière, J. Chevallier, N. de Montricher, F. Ocqueteau (eds) L’État à
l’épreuve des sciences sociales: La fonction recherche dans les administrations sous la Vème
République, pp. 295–315, Paris: La Découverte.
205
Hauchecorne, M. (2009) ‘Le “professeur Rawls” et le “Nobel des pauvres”: La

politisation différenciées des théories de la justice de John Rawls et Amartya
Sen dans les années 1990 en France’, Actes de la recherche en sciences sociales 176–7,
99–113.
Jankowki, B., Vennesson, P. (2005) ‘Les sciences sociales au ministère de la
Défense: inventer, négocier et promouvoir un rôle’, in P. Bezes, M. Chauvière,
J. Chevallier, N. de Montricher, F. Ocqueteau (eds) L’État à l’épreuve des sciences
sociales: La fonction recherche dans les administrations sous la Vème République, pp.
267–94, Paris: La Découverte.
Jobert, B. (1981) Le social en plan, Paris: Les éditions ouvrières.
libéralisme’, in B. Jobert (ed) Le tournant néo-libéral en Europe: Idées et recettes dans
les pratiques gouvernementales, pp. 21–85, Paris: L’Harmattan.
Kessler, M.-C. (2005) ‘A la recherche d’une diplomatie savante’, in P. Bezes,
M. Chauvière, J. Chevallier, N. de Montricher, F. Ocqueteau (eds) L’État à
l’épreuve des sciences sociales: La fonction recherche dans les administrations sous la Vème
République, pp. 316–62, Paris: La Découverte.
Lebaron, F. (2000) La croyance économique: Les économistes entre science et politique,
Paris: Seuil.
Medvetz, T. (2012) Think tanks in America, Chicago, IL: University of Chicago
Press.
Memmi, D. (1996) Les gardiens du corps: Dix ans de magistère bio-éthique, Paris:
éditions de l’EHESS (L’ École des hautes etudes en sciences sociales).
Mucchielli, L. (2010) ‘Vers une criminologie d’État en France? Institutions, acteurs
et doctrines d’une nouvelle science policière’, Politix 89, 195–214.
Ocqueteau, F., Monjardet, D. (2005) ‘Insupportable et indispensable, la recherche
au ministère de l’Intérieur’, in P. Bezes, M. Chauvière, J. Chevallier, N. de
Montricher, F. Ocqueteau (eds) L’État à l’épreuve des sciences sociales: La fonction
recherche dans les administrations sous la Vème République, pp. 229–47, Paris: La
Découverte.
Penissat, E. (2009) L’Etat des chiffres: Sociologie du service statistique et des statisticiens
du ministère du Travail et de l’Emploi (1945–2008), Doctoral dissertation in
sociology, Paris: EHESS-ENS.
Penissat, E. (2012) ‘Publier des “chiffres officiels”, ou les contraintes bureaucratiques
et politiques qui façonnent l’expertise d’État: Le cas des statistiques du ministère
du Travail’, Gouvernement et action publique 4(4), 45–66.
Pollak, M. (1976) ‘La planification des sciences sociales’, Actes de la recherche en
sciences sociales 2, 105–21.
Pons, X. (2008) L’évaluation des politiques éducatives et ses professionnels: Discours et
méthodes (1958–2008), doctoral dissertation in political science, Paris: Institut
d’études politiques de Paris de Paris.
206
Ponthieux, S. (2013), L’information statistique sexuée dans la statistique publique: État

des lieux et pistes de réflexion, Rapport pour la Ministre des Droits des femmes,
octobre, Paris: Institut national de la statistique et des études économiques
(INSEE).
Revillard, A. (2016) La cause des femmes dans l’Etat: Une comparaison France–Québec,
Rosental, P.-A. (2003) L’intelligence démographique: Sciences et politiques des populations
en France (1930–1960), Paris: Odile Jacob.
Rousso, H. (ed) (1986) De Monnet à Massé: enjeux politiques et objectifs économiques
dans le cadre des quatre premiers plans, 1946–1965, Paris: Centre national de la
recherche scientifique (CNRS).
Rousso, H. (ed) (1987) La Planification en crises: 1965–1985, Paris: Centre national
de la recherche scientifique (CNRS).
Rousso, H., Margairaz, M. (1987) ‘Le Plan, une affaire d’Etat? Les responsables
des commissions du Ier au IXe Plan’, in H. Rousso (ed) La Planification en crises:
1965–1985, Paris: Centre national de la recherche scientifique (CNRS).
Rueff, J., Armand, L. (1960) Les obstacles à l’expansion économique, www.
ladocumentationfrancaise.fr/docfra/rapport_telechargement/var/storage/
rapports-publics/074000508.pdf.
Serré, M. (2001) Le ‘tournant néo-libéral’ de la santé? Les réformes de la protection
maladie en France dans les années 1990 ou l’acclimatation d’un référentiel de marché,
doctoral dissertation in political science, Paris: Université de Paris 1.
Spenlehauer, V. (1998) L’évaluation des politiques publiques en France, avatar de la
planification, doctoral dissertation in political science, Grenoble: Université Pierre
Mendès-France Grenoble II.
Stiglitz, J., Sen, A., Fitoussi, J.-P. (2009) Richesse des nations et bien-être des individus,
Paris: Odile Jacob.
Tanguy, L. (2008) ‘Retour sur l’histoire de la sociologie du travail en France:
Place et rôle de l’Institut des Sciences Sociales et Travail’, Revue Française de
Sociologie 49(4), 723–61.
Terray, A. (2003) Des francs-tireurs aux experts: L’organisation de la prévision
économique au ministère des Finances (1948–1968), Paris: La Documentation
Française.
Tissot, S. (2004) ‘Identifier ou décrire les quartiers sensibles? Le recours aux
indicateurs statistiques dans la politique de la ville’, Genèses 54, 90–111.
Walliser, A. (1989) ‘Le rapport “Nora–Minc”: Histoire d’un best-seller’, Vingtième
siècle 23, 35–48.
Walliser, A. (1995) ‘Les limites stratégiques d’un rapport administratif: Le rapport
de J. Lesourne “Education et Société”’, Politiques et management public 13(1),
107–32.
207
TWELVE
Political parties and think tanks: policy

analysis oriented toward office-seeking
Camilo Argibay, Rafaël Cos and Anne-Cécile Douillet
Are political parties spaces for the production of a form of policy analysis? This
query arises especially given the ‘programmatic function’ typically attributed to
parties, a fact that raises two questions: To what extent are the political alternatives
proposed by parties and the critiques they direct to those in power formulated
in terms of public intervention? And are such plans developed using party-based
policy expertise?
The literature on political parties provides few answers to these questions, least
of all the second. This is first of all related to the way parties are understood in
the main theoretical models. A number of analyses hold that political platforms
are of only secondary importance to parties. The notion of the ‘catch-all party’
(Kirchheimer, 1966) thus leads one to think that voters more often choose a party
according to its leader than its platform and, with electoral success in mind, that it
is better for parties to stay vague and ambiguous. But other analyses, interpreting
proposals for public intervention as ‘resources and subjects of party competition’
(Coman and Persico, 2014), are not particularly attentive to the way in which
these proposals are produced. This is notably the case in approaches holding that
policy issues pre-exist parties, that they use them strategically (Downs, 1957), or
that they position themselves according to the ‘cleavage’ they represent (Lipset
and Rokkan, 1967). The same also holds for approaches placing the choice of
issues at the heart of competition between parties (Stokes, 1963), to the extent
that the construction of issues receives less attention than differences of ‘salience’,
and possibly framing, depending on the party. The ‘cartel party’ model (Katz and
Mair, 1995) not only highlights the reduced competition between parties and the
growing similarity between party options, but also emphasises the value given to
management abilities in parties that manage the state more than they represent
society. Little is said, however, about how these abilities are constituted and used.
Recent French studies of French political parties have been attentive to field
research and frequently demonstrated a critical distance from the preceding
analytical models, but they provide few answers to the question of party-specific
policy analysis. This is partly explained by the predominance of approaches
focused on the entrepreneurial and organisational dimensions of parties that follow
a paradigm ‘of Webero-Schumpeterian inspiration’ (Sawicki, 2001). It can also
be explained by an observed ‘professionalisation’ movement, understood as the
tightening and re-structuring of party organisations around the stakes of electoral
conquest. Doctrinal activities seem to have become somewhat marginalised in
209
the main French parties, to the advantage of election-strategy considerations.

This finding emerges from work on the French Parti Socialiste (PS) (Lefebvre and
Sawicki, 2006) and Parti Communiste Français (PCF) (Mischi, 2014), while work
on the Union pour un Mouvement Populaire (UMP), to the right of the political
spectrum, stresses its managerialisation (Petitfils, 2007). Such observations do
not foster interest in the construction of official positions and party platforms.1
In light of these dead-ends, this chapter gives serious consideration to the
hypothesis that political parties may pursue public policy work. This perspective is
an invitation into the black box of political parties to see the extent to which, and
under which conditions, they are spaces of production, diffusion and mobilisation
of knowledge about and for public intervention. The question is whether party
expertise on public policy exists, regardless of its impact.2
Analysis began with a survey of the literature on political parties and policy
process studies3 to assemble pertinent findings, which were supplemented by data
from ongoing research.4 The investigation focused specifically on the national
organisations of two parties: the PS5 and the UMP, which became Les Républicains
(LR) in May, 2015. The former party Rassemblement Pour la République (RPR)
was grouped with the UMP, which was formed in 2002 by the fusion of several
parties of which the RPR was the most substantial.6 Both the PS and the RPR/
UMP/LR are governing parties that have acquired positions of power at the
national as well as local levels.
This exploratory research established the hypothesis that seeking office and
seeking policy are not antithetical objectives: in fact, developments in party
rationales, like those in the profile of governing parties’ elites, are favourable to
intensifying interest in policy issues (see the first section of this chapter). Political
parties’ professionalisation nonetheless appears to have a marked effect on their
internal production of public policy expertise: party membership is marginalised
while the electoral issues and internal competition have a structuring impact (see
the second section). Last, analysis of public policy expertise production shows
that it is mainly done in the vicinity of party organisations, due to the significant
recourse to experts outside of parties and the role of think tanks (see the third
section).
Policy issues in professional political parties

The professionalisation of political parties (as defined in the preceding section)
seems to lead to a devaluation of ideas as resources that are relevant to parties
internally, but it is probably fairer to speak of the transformation of ideal work.
Indeed, the dynamics of party activism over recent decades have had a hand in
erasing ‘master narratives’, instead favouring discourse which is more focused
on policy issues. This movement was accompanied by a transformation of party
elites, today more socialised in public policy. As a consequence, questions related
to public intervention are not absent in professional political parties.
210
Political parties and think tanks
The reformulation of inter-party competition around policy issues
Electoral competition presupposes mobilising one’s ‘political family’ and setting

oneself apart from the opponent: this is what is at stake in a party’s proposed
platform. The symbolic goods that constitute this proposal may, however, be of
various natures: more or less well-developed ideological references, theories of
greater of lesser sophistication, policy proposals of variable technical complexity.
The dynamics of the history of the PS on one side and the RPR-then-UMP
on the other have fostered competition largely focused on policy issues. The
creation of the PS in about 1970 was part of the effort to capture power after a
decade of Gaullist control. Its founding was part of a political plan, both sealing
an alliance between socialists and communists and introducing a party that was
ready to govern. To this end, the ‘Common Programme of the Left’, signed in
1972, exhibits a certain degree of sophistication in its review of a large number
of public policies, while the presidential platform of François Mitterrand in
1981 contained 110 proposals. The plan to ‘open the way for socialism’ was
thus very attentive to public policy proposals. Once in power, the conversion to
monetarism and rigour made it more difficult to mobilise ‘master narratives’ on
the excesses of capitalism and re-enforced a sector-based approach to problems,
especially as the opposition on the right then had more difficulty disqualifying
the PS as the ‘harbinger of collectivism’: ‘the controversy changes levels, and
henceforth the challenge is to the socialists’ ability to ensure the management
of the republican State’ (Jobert and Théret, 1994, 60). The neoliberal rhetoric
appropriated by the main leaders of the right in the early 1980s (Denord, 2004)
to better distinguish themselves from the left in power was also toned down: their
return to government in 1986–88 heralded a less doctrinaire and more managerial
approach to public problems.
The lesser salience of socialist or neoliberal references does not necessarily
mean a move away from ideology – for proof one need only look to Nicolas
Sarkozy’s 2007 presidential campaign, radicalising entire areas of discourse in the
UMP (Haegel, 2012). On the other hand, it means that from the latter 1980s,
party activism has been more focused on ‘competition over the issues’, especially
the environment, public security, unemployment and immigration. The issue
of the environment, gradually taken on by the dominant parties, is a good case
in point (Persico, 2014). Parties make particular efforts to pre-empt the casues
of rival organisations. On the right, for example, the UMP largely cornered the
traditional themes of the Front National (especially immigration), and in 2007
it drew from the repertoire of the left on the theme of employment policy; on
the left, the PS has positioned itself since the mid-1990s on the issues of public
security (Cos, 2012a). Especially in the context of electoral campaigns, parties
often try to shape the agenda of debates around their priority issues and policy
proposals (as with the 35-hour work week for the PS in 1997, and the UMP’s
‘fiscal shield’ in 2007).
211
The concentration on particular policy issues has been accompanied by the

mobilisation of more technical weapons in the competition between parties.
Putting figures to the cost of programmes is one example. Although some
programmes had their costs evaluated as early as the 1970s, costing proposals
became more systematic during the most recent presidential elections (Lemoine,
2008). The public debt being made into a major problem is likely part of this
trend. Moreover, if personalisation and the simplification of political discourse
are the media influences that usually come to mind, there is also an observable
media injunction for technical credibility in electoral platforms, especially where
the budget is concerned.
The forms taken by electoral competition, particularly in parties such as the
UMP and the PS, thus leave room for the development of policy analysis in the
service of political parties, due to the prominence of debates on policies to be
pursued and the means to be mobilised to that end. This movement is paired
with party leaders’ increasingly strong socialisation in public action.
Transformations of the political occupation in governing parties
Interest in policy-related questions in political parties is also prompted by on-going

transformations in political legitimacy, particularly the inclusion of public policy
‘skills’ among legitimising criteria for elected officials. It is thus important to note
that since the 1960s economic skills have gradually been converted into political
capital. While this skill was previously primarily associated with the stigmatised
figure of the technocrat, judged to be little compatible with the democratic
ideal, today it has become a key marker of political identities (Dulong, 1996).
This economic skill has consequently been very influential in styling political
entrepreneurs, both at the local (Le Bart, 1992) and national levels (Gaïti, 1990).
This symbolic construction work on political identities comes with a noticeable
reformulation of the profiles of party elites, which are increasingly distinguished
by claims to skills in economic matters and/or state expertise. High-ranking civil
servants have thus become more present in party leadership. The legitimacy linked
to having studied at a prestigious institution for the training of high-ranking
civil servants moreover tends to substitute for that traditionally associated with
the electoral career: thus, in the PS, the first generation of École Nationale
d'Administration (ENA) graduates7 also possessing significant political party
capital was succeeded by a second generation in the 1980s that acceded directly
to ministries under the left upon graduation, without having held any office
previously (Lefebvre and Sawicki, 2006). Parties’ leadership bodies also solicited
high civil servants: Mathieu Fulla (2012) has emphasised how in the 1970s the
PS actively sought to recruit high-ranking bureaucrats from the Ministry of
Finance such as Alain Boublil and Jacques Attali. The non-negligible proportion
of deputies that attended power-schools is another illustration of what a resource
it is to possess this capital: among deputies in office in 2002–07, 14.5 per cent
were graduates of Sciences Po Paris, 6 per cent were ENA graduates, and 4.5
212
per cent from another Grande École, the upper tier of selective universities (Costa
and Kerrouche, 2007). Having graduated from a Grande École additionally
facilitates access to the more prestigious parliamentary commissions (the Legal
and Finance Commissions in particular). Beginning in the 1970s, the profiles of
the highest-ranking executive branch officials were often also distinguished by
having attended a Grande École (ENA, Polytechnique), be they Presidents of the
Republic (V. Giscard d’Estaing, J. Chirac, F. Hollande) or Prime Ministers (L.
Fabius, M. Rocard, E. Balladur, A. Juppé, L. Jospin).
This rising power of expert legitimacy among politicians developed in parallel
with another movement, related to political alternation: indeed, since 1980 the left
and right have succeed one another in power at an unprecedented rate, winning
four and five national elections, respectively. This provides leaders of the PS and
the RPR/UMP regular access to elected office as well as positions in ministerial
cabinets and the administration, making it possible to acquire public management
training through experience in party activities and holding elected office, as
opposed to education in elite civil-service institutions. As Carole Bachelot stresses
in her study on socialist leaders between 1993 and 2003, it is ‘indeed the political-
administrative apparatus that has, to a significant extent, trained party leadership’
(2008a, 270): over this period, 40 per cent of the members of the party’s national
office had held a position on a ministerial cabinet. In addition to these national
movements, local elected officials (who are numerous in the parties studied here)
have become increasingly important.8 They are in direct contact with public
policy issues, especially as the decentralisation reforms begun in 1982 have re-
enforced power and fostered significant policy developments at the local level.
So it is that a substantial proportion of party elites accumulated a panoply of
public policy knowledge and know-how while attending power-schools and/
or holding a bureaucratic or elected position. Moreover, regular participation in
the sites of power – as elected official, cabinet advisor, or high-level civil servant
– allows political entrepreneurs to develop networks of expertise within the
state apparatus or business world. For the majority of elites in governing parties,
then, political professionalisation (as the rationalisation of political work in view
of electoral imperatives) is accompanied by the accumulation of dispositions and
resources favourable to the development of public policy expertise within the
party.
Intra-party policy analysis influenced by presidentialisation

Counter to a simplistic vision of political work as purely the conquest of votes,
the preceding text would seem to indicate that party elites are dealing with public
policy issues even when not in elected office, and that these issues are valuable
to inter-party competition. How this translates into party operations remains
to be examined. Two complementary characteristics can be emphasised in this
regard: the tension between how the party functions as a group and the individual
213
investment of political entrepreneurs, and how work on public policy issues fits
into the rationalised quest for elected positions.
The fragility of the bodies of experts within political parties
Since the 1970s, party-based circles of expertise have become institutionalised. In

the PS, following the 1972 creation of an ‘expert committee’, 15 or so thematic
commissions were established (and statutorily recognised in the internal rules and
regulations of 1978) and placed under the responsibility of a ‘National Secretariat
for Studies’ (Bachelot, 2008a; Sawicki, 1998). These events, resulting from
the aforementioned strategy to renew the party, helped ensure the continued
existence of these bodies of experts within the party. The number of national
secretariats proliferated significantly: in 1979, the PS had 16 national secretariats,
only half of which addressed a public policy issue. In April 2015 there were
56, 40 of which focus on a specific theme – more than the number of state
ministers and secretariats. In early 2015, the UMP9 had 30 thematic national
secretariats (along with the 12 ‘functional’ secretariats) and LR had 58 (along
with the nine ‘functional’ secretariats) in March, 2016.10 Regardless, although
these developments account for the attention political parties have given to a wide
range of topics related to public intervention, they do not allow identification of
parties’ preferred channels for shaping their public policy positions.
In practice, the activity of these commissions seems quite unequal. It depends
first on the positions held by the parties. It is logical that these commissions
should take the form of spaces for policy analysis, because they are seen as sites
for producing counter-expertise to challenge the governing majority’s decisions.
This logic relies on two complementary factors: the party’s collective interest in
publicly demonstrating its opposition to policies implemented by the party or
parties in power, and the more individual interests of those who, party member or
not, are invested in its commissions, and may harbour hopes of being ‘discovered’
and included in future ministerial cabinets (Bachelot, 2008b; Zittoun, 2001). The
counterpart of this logic is that these commissions are often largely abandoned
when the party finds itself in power, when the parliamentary arena and to a greater
extent ministerial cabinets seem like more determinant scenes for orienting the
debate on a variety of issues (Bioy et al, 2015).
The output of national secretariats also largely depends on the personal interests
and party capital of those connected with them. In the first place, an elected
official’s specialisation in a public policy issue may contribute to making the
relevant commission more active. This logic is typical of young elected officials
who are given their first national secretariat. Their strategy often consists of
creating credibility for their position by heavily investing themselves in the field
leadership assigned them, which may be more or less related to their academic
and professional training. These young elected officials are thus often engaged
in expanding the number of active members in these commissions, organising
periodic events and publishing press releases and in-depth reports (Cos, 2012a;
214
2012b). In the second place, the renown of an internal party leader may also
contribute to making these secretariats more appealing, as this fame often intersects
with the legitimacy of the theme in question (for example, the national secretariats
for the economy are often held by prominent figures11). So although there are
some favourable conditions for these bodies’ work, the position they hold remains
structurally unstable and fragile.12 Beyond the effects mentioned earlier, to which
the regular rotation of terms in office should also be added, the circulation of
work that could be done there is determined by several constraining factors that
are particularly linked to competition internal to the parties.
Policy analysis shaped by intra-party competition
Internal competition is often manifest in the establishment of advisory bodies.

Although the PS has seen the activities of its various currents focus on the
promotion of their leaders since the 1990s, there are still regular demonstrations
of concern for the prolonged examination of particular public problems. Benoît
Hamon’s ‘La Forge’ (the Forge), Jean-Louis Bianco’s ‘Pôle idées’ (Idea Centre),
Ségolène Royal’s association ‘Désir d’avenirs’ (Futures Wanted), and Martine
Aubry’s ‘coopérative d’idées Renaissance’ (Renaissance ideas cooperative) are some
recent examples. Lacking work on the subject, one might hypothesise that within
the RPR and UMP, the weaker institutionalisation of currents has limited the
development of stables of competing expertise; clubs are sometimes established
around major figures, but they do not seem to develop genuine public policy
expertise.13
Beyond that, the formation of ad hoc advisory groups on platform and/
or public policy issues is always associated with the individual initiative of the
organisation’s most central actors. From his accession to the leading position
of the UMP in 2004, Nicolas Sarkozy depicted himself as candidate for the
next presidential election and organised teams to supply him with ideas: a small
exploratory committee at first, while at the Ministry of the Interior, then a more
fleshed-out team gathered around Emmanuelle Mignon,14 who Sarkozy solicited
for the development of his presidential plan. Mignon ultimately joined the UMP
hierarchy, becoming Director of Studies in November 2004. She then got most of
the staff of this division to leave in order to replace them with young Grande École
graduates (Confavreux and Lindgaard, 2007). These ‘experts’ and ‘intellectuals’
were expected to help cement the worldviews held by Sarkozy and his team,
both in terms of diagnosing actuality (including the evaluation of implemented
policies) and making recommendations. Their work was productive, leading to
thematic conventions (on education, immigration and so on) and trips to other
countries to see ‘what is done elsewhere’, all generating reports. With an eye to
the 2012 presidential election, the PS established the Ideas Centre (Laboratoire des
idées) in 2009, intended to counter the right’s ideological offensive during the
2007 presidential election. If, despite the production of 20-some thematic reports,
the Ideas Centre’s ability to contribute to the party platform seems limited, this
215
is partly because it appeared to be an initiative of Martine Aubry (leader of the

party at the time, but defeated by François Hollande in the primaries), and the
legitimacy of its status as organ of the party was heavily contested.
Moreover, although the ‘permanent’ thematic commissions officially participate
in platform-building in pre-electoral periods, the presidentialisation at work
in both parties makes broad use of their output. Although not recent, this
development is especially obvious in the case of the PS. The platform production
chain can be divided into two basic parts. First, a team in charge of preparing
the text consults the national secretariats. Once drafted, the text is submitted
to the National Bureau for deliberation and arbitration, then confirmed by
party members, who are generally kept at a distance during this first phase. The
candidate is chosen in the second phase. The candidate and his or her team15
then largely rework the first version of the text to propose a second ‘platform’ or
‘plan’ (where a personalisation of the choice and the formulation of the proposals
appears) and to structure the campaign strategy (by the selection of priority issues);
this logic is radicalised when the candidate has come through the executive
branch. When the candidate’s inner circle only assembles a few people, some of
whom may not be party members, electoral campaigns are also revealing of the
personalised character of expertise mobilisation in parties.
Party-specific policy analysis thus takes place both in the service of and in
reaction to electoral imperatives, within a frictional juxtaposition of networks
that are simultaneously collective and individual. It is consequently caught up in
the interplay between political currents, which use it without fully controlling it.
Policy analysis from extra-party sources

Understanding policy analysis production in party organisations requires not
limiting oneself to what happens within the boundaries of a party. Indeed, such
work is largely the purview of experts who are only rarely permanent employees
or members of the party, although they are never very far from it. Engaged in
more or less formal bodies, such people contribute directly to the elaboration
of political proposals.
How parties use expertise
The professionalisation of political parties results in an organisational division

of labour. The elaboration of political ideas and solutions is more the business
of experts than it is of activists. Among these experts are ‘organic intellectuals’
who are in charge of studies and training for party leadership bodies. They
may, like Henri Weber and Alain Bergounioux in the PS, be in charge of the
party’s intellectually oriented journal. Their role is nonetheless often limited to
establishing and maintaining relations between party leadership and experts in
specific sectors.
216
The experts called upon by parties are primarily from two professional worlds
that have historically been competitors in French public expertise (Barbier et al,
2013). The first, rather typical of the French political field, is that of high-level
civil servants. A strong technocratic expertise developed in France with the
advent of the Fifth Republic, even within administration (Dulong, 1998). These
high-ranking civil servants are rather easily mobilised by party leadership, a great
many of whom, as mentioned earlier, attended selective civil service schools. This
category of experts is often found in the most prestigious commissions (typically
economic). Their experience in upper administration manifests in an ability to
produce briefs and give a highly operational dimension to their recommendations.
Based on ‘technical’ legitimacy, this expertise is nonetheless not politically neutral
and, for example, has contributed to the circulation of neoliberal approaches
in – or, more precisely, at the head of – the PS, as early as the 1970s (Jobert and
Théret, 1994).
In addition to high civil servants, party authorities call on academics. They are
particularly solicited for ad hoc work tied to platform development, as illustrated
by invitations from the Ideas Centre or Mignon’s preparation of Sarkozy’s
campaign. Work meetings thus resemble research seminars, akin to what is familiar
to academics in their usual professional setting, but always with a ‘requirement for
results’ (Chahsiche, 2014, 92), meaning an imperative for specific proposals. These
academics are mobilised for their broader knowledge of the issues at hand, but
most of all they confer a more intellectual dimension to the resulting expertise,
surpassing the register of isolated solutions. Economists are mobilised the most
often, but jurists, sociologists and political scientists also participate in such work.
Although both kinds of expert can be found in a variety of bodies of experts in
service to a given party, their mobilisation varies according to the circumstances:
academic expertise seems compatible with the middle-range work inherent in
party-related organisations (commissions, research groups and so on), while high
civil servants seem to operate in a way that is more suitable to election time.
Writing briefs containing assessments of a situation, avenues for recommendations,
and some ‘language elements’ is an exercise familiar to senior civil servants that is
objectively adapted to campaigning. Indeed, they are processes that are as relevant
to the vagaries of the media agenda (ordering candidates to react on a variety
of topics) as they are the occasion for projecting into potential state governance
(assessments of overall accounting and budgetary balances, costing proposals,
evaluating their legal feasibility and so on).
By relying on these two professional profiles of people, whose contributions
may be individual or collective, the PS and UMP mobilise expertise outside
organisational boundaries. These experts often (but not always) have political
sympathies for the party which they are helping, as demonstrated in its absence
by the refusal of some of the experts solicited by Mignon because they did not
identify with the UMP (Confavreux and Lindgaard, 2007). The same logics
are found in more enduring organisations that gravitate around parties and also
contribute to the production of policy analysis: think tanks.
217
The role of think tanks alongside political parties
The role played by certain French think tanks reveals the existence and assertion
of bodies specialised in the production of ‘political solutions’ alongside political
parties. Most of these think tanks maintain a privileged but nevertheless ambiguous
relationship with a particular political party: they may provide expert advice and
political proposals to a given party under certain circumstances, and bypass it in
others. Although they are formally distinct entities, think thanks and political
parties may in some respects be interconnected through the people involved and
their networks.
Think tanks, as professionalised non-profit organisations producing public
policy-related expertise and claiming a degree of independence (Stone, 2000),
emerged rather late in France, where this kind of organisation began to proliferate
around the year 2000. French think tanks have much more limited resources
(financial and human) than their British or American peers, and their political role
is considerably less institutionalised (Medvetz, 2012). Although party affiliations
are less strong than with German foundations (Dakowska, 2004), the main
generalist think tanks in France have clearly identified connections with a political
party (Schmid, 2013): Fondapol and the Institut Montaigne are close to the UMP,
Terra Nova and the Fondation Jean Jaurès to the PS, and the Fondation Gabriel Péri
to the PCF. This proximity results from a significant circulation of individuals
between the two organisational types: think tank founders are often members of
political parties, think tanks approach elected officials as experts, and think tank
experts and directors frequently assume political responsibilities (as is the case at
Terra Nova and Fondapol). In addition, some think tanks are specialised in particular
sectors, devoted to a specific issue (international relations, the environment, health
for example),16 and less directly connected to parties. To do their work well,
think tanks turn to sector experts, often recruited from various edges of the field
of power (high-ranking civil servants, academics, consultants, business people).
Following political parties’ example, there are few representatives from union
and associative worlds among these experts, at least in generalist think thanks.
Named as individuals or as part of thematic groups, these experts work on the
development of various kinds of documents (briefs, reports, books, interviews,
on-line videos and the like) published and distributed by the think tanks. Despite
this formal diversity, the proffered arguments have much in common, such as
evaluations of implemented policies, international comparisons, references to
other studies and reports, and costing and legal counsel concerning the proposals.
One particular circumstance where think tanks work on producing policy
proposals is during the preparation of electoral campaigns. The creation of
Terra Nova, for instance, was connected to plans for intellectual production in
the service of a party: a small group of high-ranking civil servants, intellectuals,
and young entrepreneurs began meeting in 2007, after Ségolène Royal’s defeat
in the presidential election, and created Terra Nova in 2008 to ‘respond to the
ideological deficit’ of the PS, looking ahead to the 2012 elections. Anticipating
218
this electoral deadline, Terra Nova worked to get experts to participate in thematic
groups, which met over the course of several months and published reports at
the conclusion of their work. Consequently, 32 reports were published between
February 2008 and May 2012, of which 28 concern public policy and the other
four address ‘political strategy’ (Argibay, 2016). This policy analysis work is done by
an organisation that claims to be distinct from the party, but which contributes to
reflection on its platform; the think tank thus presents itself as intellectual support
for the party and its candidate, which have the final say over which elements will
end up in the electoral platform.
Fondapol’s positioning in 2012 shows that under certain circumstances even a
think tank closely connected to a party17 may work at a distance from it. At the
time, Fondapol made no claim to being in the UMPs service: the foundation’s
General Secretary, Dominique Reynié, said he wanted to have an influence on
the debate and raise themes, but without working directly for the UMP. This is
notably explained by the previously described parallel undertakings of the UMP’s
presidential candidate and the competition internal to the party. This example
attests to the fact that think tanks may position themselves as being involved in
political competition alongside political parties, with intellectual outputs that are
partly in competition with theirs.
Think tanks continue to produce policy solutions outside of periods of electoral
campaigning. Since 2012 and the arrival of the PS in power, Terra Nova continues
to produce expertise on state-level public intervention without going through
the party. It is no longer concerned with contributing proposals or ‘ideas’ to a
party, instead intervening more directly with those in power. In addition to the
publication of briefs and reports, its recommendations are disseminated through
informal meetings with members of the administration and ministerial cabinets
as well as through presentations at the National Assembly. It is sometimes difficult
to tell if it is a matter of providing proposals or organising forms of governmental
communication; Fondapol’s thematic conferences in the early 2010s, which
were regularly opened by a minister in the Fillon administration, are a case in
point. Whatever the case, the relationship is not built with a party, but with a
governmental team. Think tanks also act apart from parties by publishing most
of their results (digitally or in print), thus distributing their ‘political solutions’
to interested readers. Last, beyond relationships with public leaders, some may
also take on a cause; this was the case of the Institut Montaigne, which in 2004
was the first organisational actor to represent the theme of ‘diversity’ (Bereni,
2009) in public debate.
Conclusion
Analysis of political parties’ contributions to the production of knowledge on
and for public intervention is a topic with considerable potential for further
study. The issue has thus far been neglected in work on political parties, which
predominantly looks at the professionalisation process through the renewed
219
focus on electoral issues. Regardless, this chapter has shown that this process
is not contradictory to ideal political party work, even if the ways this work is
accomplished take on characteristics of professionalisation. Activists are thus kept
at a distance from such activities, at best limited to the validation of the electoral
platform; this work is party leadership business, and more specifically that of the
experts surrounding them.
Examining political parties through the question of party-specific policy analysis
thus makes it possible to see what party professionalisation conceals in terms of
ideal work. It is particularly typified by intensification during electoral periods and
a strong division into sector-specific work, as demonstrated by the multiplication
of national secretariats at the head of both of the studied parties and the thematic
organisation of the think tanks close to them. Intra-party policy analysis is not
about working on overarching ideologies, but focuses instead on positioning the
party on the issues and producing political solutions for implementation.
Another aspect of this relationship between political professionalisation and
ideal party work concerns the sites where this work is produced. Indeed, research
highlights that these bodies are less often truly the party’s bodies of experts than
they are more or less formalised spaces associating people from the party and an
assortment of other agents typically able to boast of academic titles and expert
resources. Far from putting these separate worlds into conflict, these spaces show
the mutual acquaintanceship connecting a substantial proportion of party elites to
state expertise, especially from ministerial divisions. In this sense, study of French
governing parties from the perspective of their public policy expertise re-enforces
the sense of a strong interpenetration between governing parties and the state.
Notes
1 Other parties, such as the extreme-right Front National, have inspired research on the
construction of their ideological positions, but it has remained marginal (Dézé, 2006). It is also
noteworthy that French political science has recently seen a surge in interest in the topic of
‘platforms’ (see Bué et al, 2015).
2 The question of parties’ influence on public policy, the core of several studies (see the Comparative
Agendas Project), is not of direct interest here.
3 Policy studies also have little to contribute to this analysis, especially because of the secondary
role often attributed to political actors in public policy-making (Sawicki, 2002; Douillet and
Robert, 2007). The rare studies raising this question (Jobert and Théret, 1994; Zittoun, 2001)
emphasise the limited or even non-existent expertise within the party in question.
4 Rafaël Cos is conducting doctoral research on the production of electoral platforms in the Parti
Socialiste between 1995 and 2012, and Camilo Argibay holds a post-doctoral position researching
think tanks in France.
5 The Parti Socialiste was formed under this name in 1969, taking over from the SFIO (Section
Française de l’Internationale Ouvrière), and integrated several other parties identifying with socialism
in 1971.
6 The RPR was founded in 1975.
7 The École Nationale d’Administration (ENA) is the most selective and prestigious civil service
school, and its graduates, known as énarques, are considered to form a networked national elite.
8 In the PS, sometimes characterised as ‘the party of elected officials’, 40 per cent of members
were local elected officials in 2011. Additionally, 8 per cent of members worked with elected
220
officials, and 13 per cent were employees of regional, departmental, or local-level governments
(mostly in middle and upper management), further re-enforcing the weight of members working
on local public policy issues (Dargent and Rey, 2014).
9 To the authors’ knowledge there is no published work on the history of bodies of experts in
right-wing parties.
10 The themes include sports, education, tourism, justice, health, disabled persons and public
security.
11 Significantly, in the PS, the National Secretariat for the Economy was assigned to a woman
once only (Karine Berger, in 2012), although women are quite numerous to hold a position
in the national secretariats. The key position of head of the National Secretariat for Studies has
never been assigned to a woman either since 1971.
12 In her study, C. Bachelot (2008a) also mentions the PS’s Economic, Social, and Cultural
Committee. Supposed to take over from the ‘group of experts’ after 1998, and more open to
the participation of members of ‘civil society’, this committee seems nonetheless marked by
the limited attention it receives within the party, and the selected themes could also be seen as
relatively ‘minor’ (time of life, urbanism, associative life and so on).
13 F. Haegel (2012) thus evoked the existence in 2009 of ‘Dialogues et initiatives’ (Dialogues and
Initiatives), ‘France.9’, and ‘Le chêne’ (The Oak), formed around J.-P. Raffarin, F. Fillon and
M. Aliot-Marie respectively, and ‘Génération France’ (France Generation) for J.-F. Copé. She
describes these ‘advisory clubs’ as being formed mainly for the promotion of a particular figure.
14 A graduate of a prestigious business school (ESSEC) and an exclusive political science institute
(IEP de Paris), Emmanuelle Mignon was first in her class at ENA in 1995 and quickly joined
the Conseil d’Etat (the supreme administrative court). She was named cabinet secretary under
Nicolas Sarkozy in 2002.
15 The composition of these teams is an indicator of the marginalisation of women in party-based
ideal work. As for the PS, only five women have had a meaningful position in the teams in
charge of electoral platforms, three of them being implied in the elaboration of Segolène Royal’s
platform for the 2007 presidential election. The key role played by Emmanuelle Mignon for
Nicolas Sarkozy’s platform in 2007 appears to be an exception in the gender division of labour
within political parties.
16 This is the case for the Fondation Nicolas Hulot for environmental issues, for example, or the
Foundation for Strategic Research (Fondation pour la Recherche Stratégique) for international issues.
17 Fondapol was founded in 2004 with the support, including financial, of the UMP.
References
Argibay, C. (2016) ‘La construction d’un projet politique par fragments: Le
rôle du think tank Terra Nova pendant la campagne de 2012’, in J. Gervais,
G. Courty (eds) Enjeux présidentiels: Groupes en campagne (2012) Lille: Presses
Universitaires du Septentrion.
Bachelot, C. (2008a) ‘Groupons-nous et demain’: Sociologie des dirigeants du Parti
socialiste depuis 1993, doctoral dissertation in Political Science, Paris: Sciences
Po Paris.
Bachelot, C. (2008b) ‘Revisiter les causalités de l’évolution: Le PS au prisme
de la cartellisation’, in Y. Aucante, A. Dezé (eds) Les systèmes de partis dans les
démocraties occidentales, pp. 384–410, Paris: Presses de Sciences Po.
Barbier, M., Cauchard, L., Joly, P.-B., Paradeise, C., Vinck, D. (2013) ‘Pour
une approche pragmatique, écologique et politique de l’expertise’, Revue
d’anthropologie des connaissances 7(1) 1–23.
221
Bereni, L. (2009) ‘Faire de la diversité une richesse pour l’entreprise’, Raisons

politiques, 3(35) 87–105.
Bioy, X., Eymeri-Douzans, J.-M., Mouton, S. (2015) Le règne des entourages:
Cabinets et collaborateurs de l’exécutif dans la France contemporaine, Paris: Presses de
Sciences Po.
Bué, N., Fertikh, K., Hauchecorne, M. (2015) Les programmes politiques: Genèses
et usages, Rennes: Presses universitaires de Rennes.
Chahsiche, J.-M. (2014) ‘De l’ “éthique du care” à la “société du soin”: La
politisation du care au Parti socialiste’, Raisons politiques 4(56) 87–104.
Coman, R., Persico, S. (2014) ‘Politiques publiques et partis politiques’, in L.
Boussaguet, S. Jacquot, P. Ravinet (eds) Dictionnaire des politiques publiques, pp.
482–90, Paris: Presses de Sciences Po.
Confavreux, J., Lindgaard, J. (2007) ‘L’hémisphère droit: Comment la droite est
devenue intelligente’, Mouvements 4(52) 13–34.
Cos, R. (2012a) ‘Le Parti socialiste et la régulation politique des désordres:
L’émergence d’un nouvel entrepreneur de sécurité’, in M. Boucher, V.
Malochet (eds) Regards croisés sur la régulation sociale des désordres, pp. 75–90,
Paris: L’Harmattan.
Cos, R. (2012b) ‘Les élus socialistes face aux chiffres de la délinquance:
Dispositions, positions et prises de positions partisanes sur les statistiques
policières’, Mots: Les langages du politique 100: 107–22.
Dakowska D. (2004) ‘Les fondations politiques allemandes en Europe centrale’,
Critique internationale 24(3), 139–57.
Dargent, C., Rey, H. (2014) ‘Sociologie des adhérents socialistes: Rapport
d’enquête’, Les cahiers du CEVIPOF, 59 (entire issue).
Denord, F. (2004) ‘La conversion au néo-libéralisme’, Mouvements 5(35), 17–23.
Dézé, A. (2006) ‘Le Front National comme “entreprise doctrinale”’, in F. Haegel
(ed) Partis politiques et systèmes partisans en France, pp. 255–84, Paris: Presses de
Sciences Po.
Douillet, A.-C., Robert, C. (2007) ‘Les élus dans la fabrique de l’action publique
locale’, Sciences de la société 71, 3-26.
Downs, A. (1957) ‘An economic theory of political action in a democracy’, The
Journal of Political Economy 65(2), 135–50.
Dulong, D. (1996) ‘Quand l’économie devient politique: La conversion de la
compétence économique en compétence politique’, Politix 9(35), 109–30.
Dulong, D. (1998) Moderniser la politique: Aux origines de la Ve République, Paris:
L’Harmattan.
Fulla, M. (2012) Le Parti socialiste face à la question économique (1945–1981): Une
histoire économique du politique, doctoral dissertation in Political Science, Paris:
Sciences Po Paris.
Gaïti, B. (1990) ‘Des ressources politiques à valeur relative: Le difficile retour
de Valéry Giscard d’Estaing’, Revue française de science politique 40(6), 902–17.
Haegel, F. (2012) Les droites en fusion: Transformations de l’UMP, Paris: Presses de
Sciences Po.
222

libéralisme’, in B. Jobert (ed) Le tournant néo-libéral en Europe: Idées et recettes dans
les pratiques gouvernementales, pp. 21–85, Paris: L’Harmattan.
Katz, R., Mair, P. (1995) ‘Changing models of party organization and party
democracy. The emergence of the cartel party’, Party Politics 1(1), 5–28.
Kirchheimer, O. (1966) ‘The transformation of the Western European party
systems’, in J. LaPalombara, M. Weiner (eds) Political parties and political
development, pp. 177–200, Princeton, NJ: Princeton University Press.
Le Bart, C. (1992) La rhétorique du maire entrepreneur: Critique de la communication
municipale, Paris: Pedone.
Lefebvre, R., Sawicki, F. (2006) La société des socialistes: Le PS aujourd’hui,
Bellecombe-en-Bauges: Editions du Croquant.
Lemoine, B. (2008) ‘Chiffrer les programmes politiques lors de la campagne
présidentielle 2007: Heurs et malheurs d’un instrument’, Revue française de science
politique 3(58), 403–31.
Lipset, S., Rokkan, S. (1967) Party Systems and Voter Alignments: Cross-National
Perspectives, New York: Free Press.
Medvetz, T. (2012) Think Tanks in America, Chicago, IL: University of Chicago
Press.
Mischi, J. (2014) Le Communisme désarmé: Le PCF et les classes populaires depuis les
années 1970, Marseille: Agone.
Persico, S. (2014) Un clivage, des enjeux: Une étude comparée de la réaction des grands
partis de gouvernement face à l’écologie, doctoral dissertation in Political Science,
Paris: Sciences Po Paris.
Petitfils, A.-S. (2007) ‘L’institution partisane à l’épreuve du management’, Politix
3(79), 53–76.
Sawicki, F. (1998) ‘The Parti Socialiste: From a party of activists to a party for
government’, in P. Ignazi, C. Ysmal (eds) The Organization of Political Parties in
Southern Europe, Wesport, pp. 70–8, London: Praeger.
Sawicki, F. (2001) ‘Les partis comme entreprises culturelles’, in D. Cefaï (ed) Les
cultures politiques, pp. 191–212, Paris: Presses universitaires de France.
Sawicki, F. (2002) ‘Du parti à l’assemblée régionale: Itinéraire d’un professionnel
de l’intermédiation’, in O. Nay, A. Smith (eds) Le gouvernement du compromis,
pp. 23–45, Paris: Economica.
Schmid, L. (2013) ‘Pourquoi les partis ne sont pas producteurs d’idées?’, Esprit
8–9, 40–2.
Stokes, D. (1963) ‘Spatial models of party competition’, American Political Science
Review 57(2), 368–77.
Stone, D. (2000) ‘Non-governmental policy transfer: The strategies of independent
policy institutes’, Governance 13(1), 45–62.
Zittoun, P. (2001) ‘Partis politiques et politiques du logement, échange de
(October).
223
THIRTEEN
Economic interest groups and policy

analysis in France
Guillaume Courty and Marc Milet
Knowledge production in and about public policy by economic interest groups in

France – that is, the professional groups that represent and defend the economic
sector and its affiliated companies – has given rise to almost no specific studies.
Consequently, although historical (Guillaume, 2004; Lefranc, 1976) and
sociological (Offerlé, 2013) studies1 abound on ‘employer organisational space’,
knowledge of the conditions of the production of public policy analysis conducted
by major employers themselves remain vague. This is undoubtedly linked to
the paucity of studies on political economy and industrial relations, and to the
existing division between the approaches used in the analysis of public policy on
the one hand and the analysis of interest groups on the other. The portmanteau
word ‘expertise’ has simply been a point of convergence (Restier-Melleray, 1990).
Policy analysis has thus been primarily and implicitly perceived as knowledge
manipulated by policymakers to legitimise public policies, or as one of the sets
of action used by interest groups (Robert, 2008).
This chapter focuses on three areas: first, it analyses organisations in order to
identify and highlight the types of analytical structures in place; second, it analyses
selected texts from sectors from which our field study was conducted and for
which we have sufficient reliable data (for instance, we have very specific data on
transport and small and medium enterprises); third, it analyses the documentation
issued by organisations within the context of the 2012 presidential elections. We
have decided against proposing an overview of doctrinal positions and instead
we have chosen to focus on how analytical approaches and structures have been
developed.
We defend two key ideas: first, although there is a long-held perception of
France as a dominant state which tends to impose its expertise, a presentation of
the traditional co-production of encrypted data reveals a lasting and more complex
relationship between economic interest groups and the French government. In
this sense, even during the ‘golden age of the State’ (Courty and Suleiman, 1997),
the authorities long depended on the data and studies provided by economic
interest groups. Moreover, the process of producing standards meant that policy
analysis conditions varied greatly from one type of text to another – that is,
regulatory or legislative – and across sectors. This puts the perception of an ad
hoc, unequivocal and national French model into perspective. The first section
presents the latest developments and the groups involved in policy analysis. The
second section explores the different analysis modes within a context that revisits
225
the interventionist state model. We seek to situate, over the long term, how the
different forms and modes of analysis produced by interest groups have been
organised. These forms and modes reveal early professionalisation in a real lobbying
know-how à la française which conflicts, however, with the discourse promoted
by economic actors themselves. These actors argue that the assumed delay of the
French in practising a logic of influence is the result of a philosophical and political
tradition hostile to intermediate bodies that has influenced the country ever
since the French Revolution. The third section analyses the expertise developed
in public policy and presents the different proactive and reactive conditions of
policy analysis. The last section presents the current developments in the context
of state transformation.
Little-known economic interest group activities

Public policy analysis by economic interest groups has remained an unexplored
area of research; indeed, emphasis has been placed on renewed analyses around
economic interest groups on the one hand and public policy studies on the other.
We must therefore begin by questioning this underdevelopment and presenting
the actors involved.
Policy analysis by economic interest groups: an addressed but vague activity
A brief survey of studies that have focused on general analysis of economic

interest groups reveals five types of approaches that provide implicit details on the
little attention that has been paid to how these groups undertake policy analysis
activities. The interactionist approach examines the relationship between interest
groups and the state (based on the definition of a pluralistic, neo-corporatist
and interventionist model). In this context, to grasp the activities of economic
actors, public policy is analysed in terms of policy networks and through the
pursuit of a national model. Situating France has been a longstanding issue.
The agricultural and industrial sectors have been central, both in validating the
French version of the neo-corporatist model (Keeler, 1987) and in initiating a
pluralist model characterised by niches of sectoral co-management, specific to
meso-level corporatism (Wilson, 1987; Jobert and Muller, 1987). Analysis of
industrial policies and labour relations has primarily presented an opportunity
to show the privileged position of the state as an agent of regulation, placing it
on an equal footing vis-à-vis civil society (Schmidt, 1996). Irrespective of the
model chosen, there has been a tendency in the literature to implicitly consider
that policy analysis lies primarily not within interest groups, but, rather, within
the ‘public expertise’ developed by high-ranking institutions (Suleiman, 1979).
A second approach is part of the sociology of power and elites. In its French version,
these dominant actors have a functional constraint that acts in their favour (that
is, they do not hinder economic dynamics) and exempts them from working with
public authorities (Schmitter and Streeck, 1985). French literature reflects this
226
Economic interest groups and policy analysis in France
interpretation by underscoring the perpetual weakness of umbrella organisations

(the National Centre of French Employers (Centre national du patronat français
(CNPF), then the Movement for French Companies (le Mouvement des entreprises de
France [MEDEF]) and, more broadly, of the different representative organisations
in relation to firms or entrepreneurs’ clubs and think tanks. In this context, recent
studies have focused on the emergence of consultants in the policy monitoring
field (Grossmann and Saurugger, 2004; Henry and Pierru, 2012) which has been
accompanied by the integration of experts in the new structures created to reform
entire sectors (for health, see Benamouzig, 2012).
Shifting away from comparative politics and the challenges that emerge
during state and interest group interactions, sociology has adopted a qualitative
approach partly based on research monographs (Michel, 2006; Offerlé, 2013) and
longitudinal studies. This sociological trend, which has also revived the internal
and organisational approach of interest groups (Salisbury, 1969), questions the
feasibility of measuring interest group influence (Lowery, 2013) and its relevance
(Courty, 2006; Offerlé, 1998). This approach thus resonates with perceptions
of policy analysis: as the activities conducted by economic interest groups do
not merely seek to influence policy-making, the manner in which norms and
data are interpreted essentially fall within the ‘logic of membership’. This logic
publicly reveals groups’ actions and appears to be primarily oriented toward their
members. Challenging French lobbying calls for the redefinition of its outlines:
within the context of a comprehensive approach, which activities do actors
consider as policy analysis? Is the provision of a scope for public issues viewed as
a lobbying activity, a policy analysis activity, or both?
Taking into account economic interest groups when forming the ideas that
inform action programmes (Muller, 1982) is thus a fourth path through which to
initiate policy analysis. The transformation of employers’ umbrella organisations
(from CNPF to MEDEF in 1998 [Woll, 2006]) can thus be interpreted as a
manifestation of the new ambitions to produce an almost doctrinal corpus in
the quest to influence and shape the social regulatory policy model: employer
organisations propose a ‘new social constitution’ based on a radical reform that
advocates change in how the scope between the law and the contract is divided,
with the objective of including the social sphere in the contract (Duclos and
Mériaux, 2001; Palier, 2002). It was thus possible to present the positions defended
in the early 2000s by D. Kessler – then deputy vice president of MEDEF and
president of the French federation of insurance companies – as the seizure of
power by the liberal trend and analysis within the umbrella organisation.
This generic interpretation of policy analysis, however, must be interpreted with
caution insofar as it tends to use an extensive logic and integrates all the activities
of a group within this analysis. Available studies have as yet been unable to clarify
the real contribution of economic interest groups in relation to other structures
(such as think tanks). Similarly, the precise conditions in which an internal corpus
is produced (including but not limited to MEDEF) have remained inaccessible
to date. The final path through which policy analysis has been initiated has
227
consisted of considering economic actors within the local space when dealing with a
territory. Such areas, however, may show no interest in establishing relationships
with political authorities as the study of the bourgeois of Marseille (Zalio, 2004)
has shown. In this case, policy analysis is an endogenous activity that seeks to
better understand the norms and economic action framework, while keeping its
distance from the political field.
The recent interest of political science for economic interest groups
Each in their own way, the approaches mentioned above have increasingly begun
to focus on economic interest groups. The presentation below shows the density
and diversity of this area, as well as exploring some of the misconceived areas of
political science.
With the exception of the agricultural sector, French employer organisational
space is structured into seven organisations (Offerlé, 2011). Four of these are
viewed as representative organisations: MEDEF was established in 1998 to replace
CNPF, which was established in 1946 (74 organisations and 750,000 members);
the General Confederation of Employers of Small and Medium Companies la
Confédération des petites et moyennes entreprises (CPME) (ex-Confédération générale des
petites et moyennes entreprises [CGPME]), established in 1944 (92 organisations and
550,000 companies); the professional workmanship union (l’Union des entreprises
de proximité [U2P] [ex-UPA]), established in 1975 (59 organisations and 300,000
members); The National Union of Self-employed Professionals (l’Union nationale
des professions libérales [UNAPL]), established in 1977 (58 organisations and 110,000
members). In addition to these organisations, of which membership is voluntary,
there are also consular organisations of which membership is mandatory: the
assembly of French Chambers of Commerce and Industry (l’Assemblée des chambres
de Commerces et d’Industrie [ACFCI]) and Permanent Assembly of Chambers of
Trades and Crafts (l’Assemblée permanente des Chambres de métiers et de l’artisanat
[APCMA]). Last, the French association of private companies (Association française
des entreprises privées [AFEP]), established in 1982, concerns the 100 largest private
groups. Hundreds of alliances and thousands of federated trade unions have
developed from these umbrella organisations.
Public policy analysis can thus be developed at the three levels comprising
economic interest groups. At the first level are the structures, that is, companies.
Above these are the federations and unions of the different branches and sectors.
At the top level are four representative organisations, and also the three key players
in policy-making and policy monitoring. French political science has only very
recently begun to show an interest in approaches of interest groups. Moreover,
it has long placed emphasis on the top level and overlooked the other two levels.
Despite the existence of numerous studies focusing on sectoral public policies
which might be expected to fill this gap, in most cases, these studies have focused
on the top levels of the state, once again leaving aside interest group structures
involved in the formulation and monitoring of public policies. Consequently, we
228
cannot fully address this gap. At best, we can propose some generalisations based
on monographs focusing on the organisations of which we have information on
how public policy analysis is undertaken.
To provide a summary of the new configuration of which these organisations are
now part, a few recent developments must be highlighted. These developments
have affected organisations’ structures and hampered their analytical activities.
After their incorporation into the policy network that the French government
developed after the Second World War – from which the definition of a ‘planned
economy’ or ‘neo-corporatism’ emerged – economic interest groups were pulled
to one side by the European project that was launched in the second half of the
1950s involving the energy, agriculture and transport sectors. They were pulled to
the other side by the increasing power of local authorities, revived in France by the
decentralisation laws passed from 1982. The transition to the twenty-first century
was thus inspired by more than the neoliberal turn; indeed, the classification of
areas of expertise in economic policy was modified. In this three-dimensional
space – supranational, national and local – economic interest groups have now
become potential actors and companies have become legitimate actors among
these institutions. Moreover, umbrella organisations such as branch federations
have not ceased in their attempts to monopolise expertise in their sectors.
Economic interest groups and the interventionist French model

Analysis of economic interest groups can be influenced by two models. On
the one hand, it is difficult to disagree with the idea that the state dominates
in France, making it a model of a system driven by state-centred interests. On
the other, it is equally difficult to avoid addressing the business community’s
control over some public policies, a model that can also be found in British and
American case studies. With regard to the ninteenth century, these two models
must be viewed in perspective. Indeed, it is noteworthy that although economic
interest groups and the state jointly produced policy instruments and policy issues,
this activity was not as important and central as it had been up until the mid-
1960s. Subsequently, it was necessary to re-examine the traditional separation
of instruments that the government brought into play to develop and analyse
public policy. Among the most fundamental instruments were socio-economic
data, which fuelled simulation and forecasting models. Rather than representing
state knowledge as such, these data were co-produced.
The co-production of statistics and macroeconomic forecasting
Economic interest groups have always had a special role in policy analysis because,
relatively early, governments have always conferred on them the role of informers
and producers of socio-economic data. Far from being actors working outside the
government, economic interest groups have been involved in the development
of statistics, even though these data have been claimed and monopolised by
229
state agents since the 1950s. Owing to these past experiences, interest group
representatives have internalised the custom of transmitting figures that political
agents have become accustomed to expecting from them. Indeed, the social history
of statistics also pits business interest groups and government elites against each
other to understand how three tools useful for macroeconomic forecasting are
produced: the economic sector, the product and the area of activity (Courty, 2006).
From the onset of the institutionalisation of the representation of interests,
as part of the National Economic Council (Conseil national économique [CNE])
which foreshadowed the Economic and Social Council of the Environment
(Conseil Économique et Social de l’Environnement [CESE]), the available statistical
data were extracted from the population census (Volle, 1982). Surveys of different
branches and production surveys were the instruments used then (Margairaz,
1991). They helped create a link between administrative groups and offices and
built the networks necessary for quantification and for the required additional
data (Chambers of Commerce, Chambers of Agriculture and trade unions). The
inclusion of these groups was shaped by two decrees issued in 1938, obliging
employers with more than five employees to produce statistical data or face a fine.
The first directories and regular newsletters were published from 1945; while one
counted, the other informed. This task was recognised and guaranteed by the state
for 88 employer organisations in 1939, then for organisation committees created
by the Vichy government to replace the trade unions of the Third Republic, and,
last, for the remaining 235 approved professional organisations in 1966. However,
only high-ranking government officials held the keys to the formalisation of the
figures and programme development. The division of statistical activities between
interest groups and high-ranking officials was thus established and persisted with
planning.
As Volle has pointed out, this co-production required actors to go through
a learning process. Companies ‘certainly had no in-depth accounting skills’
allowing them to transmit the expected data. Moreover, it was in the best
interest of employers to fabricate lies or develop ‘administrative arrangements’
because their contributions to economic interest groups partly depended on the
data transmitted and there was thus a certain tendency to underestimate data
in order to reduce one’s contribution. Indeed, inconsistency between the data
from tax returns and data collected in surveys across different branches resulted
in the implementation of systematic industrial censuses. Co-production was not
a marginal phenomenon. Between the Liberation and the industrial census of
1962, approximately 1,000 people working for employer organisations transmitted
data that was then transformed into statistics by 15 state officials (Volle, 1982).
Once launched, National Agency of Statistics (Institut national de la statistique
et des études économiques [INSEE]) fought against the CNPF in particular, as
it sought to impose its surveys and their publication, and to organise and provide
annual data (see the annual business survey from 1968).
The co-production approach also applied to the forecasting tools used in the
context of the Marshall Plan. Moreover, it proved relevant, helping to explain
230
the sectoral policies structured from the interwar period. Here again, economic
interest groups were solicited, and some were even funded by the government.
Some of the most legitimate of these groups thus shifted from their roles as
government critics and became official informers. Between 1921 and 1923,
the Federation of Industrialists and French Traders (Fédération des industriels et
des commerçants français [FICF]), that is, the most representative group of French
employers at the time, became a collaborator and practically a Commerce, Labour
and Finance ministries’ official as it was contracted to assist in the production of
technical information (Dubos, 2002). Further analysis reveals that other places
also emerged, reflecting the different facets of the co-production of public policy.
The neoliberal state and the maintenance of public–private interdependence
The French interventionist state is governed by enlisting economic interest groups

– can the same be said of the neoliberal state and deregulated economic sectors?
Two examples drawn from sectoral policies may prove helpful in understanding
the changes introduced. They clearly show the transitions made by neoliberal
policies in France: deregulation; the transformation of public service concepts;
the primacy of the most open competition possible; withdrawal of the state as
contractor and state as shareholder. In both the transport and building sectors,
public policies were fragmented into several sections even while these sectors were
transitioning from regulation to deregulation. Although economic interest groups
maintained a close relationship with the government, these changes affected their
structures and their modes of public policy analysis.
With regard to fragmentation, these sectoral policies received an educational
component around the 1960s when vocational certificates (CAP), technical
school certificates (BEP) and vocational schools were created. A new technical
education policy was implemented with the economic interest groups of transport
and building sectors which had developed ad hoc organisations or specialised
departments. An obvious consequence ensued: economic interest groups ceased
to analyse economic policies alone and began to monitor, finance and transmit
data useful for educational policy. When lifelong training was developed, this
logic was reproduced in adult vocational training. This interdependence is
reflected in other government policies ranging from infrastructure management
to environmental standards.
The recent trend developed by the executive to propose legislation concerning
several public policies has also proved a constraint to economic interest groups.
Public policy analysis has been broken down by types of services because of
the new form of neoliberal policies. Indeed, these policy texts have become
increasingly heterogeneous, embracing several economic sectors and affecting each
economic actor. Policy analysis now requires monitoring of the overall activity
of authorities rather than simply carrying out legal monitoring on ministry and
parliamentary committees or specific regulatory bodies. This constraint has also
231
had an impact on companies, which now face greater legal uncertainty than in
the past.
Last, economic interest groups have maintained the production of educational
or critical analysis for their members, competing organisations, the media and
political officials: explaining, increasing understanding of and mobilising for and
against a newly adopted mechanism have remained vital traditions.2 This activity
is also subject to a graduated and varied approach that ranges from developing a
communication policy to awaiting a special issue of the professional journal of
the sector such as the indispensable Le Moniteur in the building sector.
After presenting the social history of statistics in France, this broad evaluation
approach calls for a detour through the history of the trade press because the
‘organs’ of economic interest groups have been among the supporters of policy
analysis. Professional journals have become independent, as demonstrated by the
journal l’Officiel des transporteurs, created in 1935 then integrated into the National
Federation of Road Transport (Fédération nationale des transports routiers (FNTR),
now affiliated with the CGPME). This journal became one of the publications of
the Lamy group after several reforms. In this variant, these publications enjoy the
ambiguity of their history: considered by some as the ‘unofficial’ spokespersons of
the organisation, they are generally in situations of interdependence with groups
and can thus hardly refuse to publicise their political assessment of an approved
bill or one under discussion. Irrespective of their reputation, they present another
opportunity to understand the diversity of professionals involved in this assessment,
acting as economic journalists belonging to public policy analysis ‘associated
rivals’ (Salisbury, 1991). Another possible alternative is to maintain an internal
publication enabling the organisation to communicate to both its members and
the political field in relation to how it evaluates a mechanism. Magazines, bulletins,
newsletters and, recently, multimedia (official sites and sites dedicated to specific
operations or situations – elections in particular) and audiovisual (Medef TV).
All these supports also allow organisations to assess how sectoral policy analysis
is used in the professional sphere.
In small- and medium-sized enterprises, the common point in the development
of representative organisation journals is a result of the desire to have a
dissemination channel designed in line with the sector rather than as a simple trade
union body (this includes La Volonté des PME for CGPME or Le bâtiment artisanal
for the Confederation of Crafts and Small Construction Enterprises (Confédération
de l’artisanat et des petites entreprises du bâtiment [CAPEB]). This trend reached its
ultimate goal for CGPME when the journal of professional organisation changed
from a trade union support journal to a more general publication, Perspectives
entrepreneurs. In the local commerce and craftsmanship sectors, these publications
were more a tool for the dissemination of policy information than for the analysis
of mechanisms. According to Olson’s logic, membership of craft unions in the
Fourth Republic and the beginning of the Fifth Republic enabled members
to take advantage of monthly updated data of the applicable price lists. Such
models disappeared when the general price regulation regime was implemented
232
in the 1980s. However, the decision-making model remained unchanged. Given

the increasing complexity of the VAT system, publications began to submit
applications with different tax rates depending on working arrangements (for
example, renovation, improvement of energy efficiency). This detour through
the broad approach to policy analysis that led to the inclusion of communication
activities also redefined its contours. While professional associations relayed policy
mechanisms, they also participated in their analysis.
Last, the group of large enterprises included those both publicly and privately
owned. These enterprises were attentive to abiding by current laws and anticipated
the reforms that were likely to have an impact on process manufacturing, the state
of labour laws and the conditions of sale within markets. Owing to the presence
of lawyers within the firms, the division of labour was initiated but few services
were responsible for relevant public policy analysis. In this group as well, this
activity was fragmented between the departments in charge of the employment
and social policy, and those whose responsibility it was to monitor taxation,
production standards and competition terms. The major innovation was the
development of public affairs directorates (Directions Affaires publiques [DAF]), the
only organisation to have government authorities as their interlocutors. Moreover,
during social audits and preparations for elections, they had the difficult task of
specifying how the company concerned was going to implement government
policies. Pressurised by national and European public policies, the enterprises were
also under pressure following the transformation of management that resulted
in the redesigning of their organisational structure and analytical work (Greenan
and Walkowiak, 2010).
Economic interest groups: public policy ‘experts’?

Interest groups are perceived as key experts recruited by institutions (for the
European system, see Saurugger, 2002), seeking to compensate for the shortfalls
of political parties when developing the different sections of their programmes
(Zittoun, 2001). In this field, a longitudinal approach enables the groups to
develop expert knowledge early in their respective fields. This knowledge is
referred to as ‘technical knowledge’ when it addresses production standards, ‘legal
knowledge’ when negotiations around the content of a given law are involved
and ‘economic’ or ‘fiscal’ knowledge when negotiations involve tax rates or hiring
incentives. The perception of expertise as ‘fashionable’ has generated a number of
myths that should be dispelled. The greatest myth suggests that every economic
interest group is an expert in the sectoral policies that affect it.
The example of small-scale businesses and SMEs reveals that the model of a
specific structure responsible for assessment and expertise is a myth. The integrated
model conforms to a synergy between specialised services that work together
and debate on the monitoring of legislative and regulatory reforms (UPA3 and
CAPEB). Organisations at the national level stand out as key institutions: on the
one hand, they can be a resource to local structures which turn to them in their
233
quest for information on policy systems; on the other, they can seek out local
representatives to participate in the working groups established by government
departments (a specific case is consultations with the department management
team when preparing decrees) or acquire ‘local reviewers’ of proposals developed
by the central authority. The rule, at CAPEB, is therefore to designate a
representative ‘project leader’ in the office who will be in charge of a theme and
work with a permanent team.
Models of expertise: competition and evolution
Transforming types of outsourcing rather than shifting from an integrated

model, strictly speaking, to an outsourced model, has historically been more
commonly observed among small-scale businesses and SMEs. From 1946 to 1972,
CGPME developed a highly personalised expertise model, notably via the use
of a business law consultant who played only an ‘advisory’ role in ‘discussions
and negotiations’ with parliamentarians and the administration, and was the
recognised representative of SME expertise within the public space. Similarly,
MEDEF created a department of public affairs chaired by a former congressman.
At the turn of the 2000s, organisations representing craftsmen and SMEs turned
to consulting firms. A hybrid organisational configuration of policy analysis
followed: far from shifting from an internal analysis model to an outsourced model
of analysis, arrangements between the two types of models were transformed. For
instance, although UNAPL had an employee responsible for policy affairs, it also
used the services of consulting firms. Consulting firms were thus called upon to
provide quantified data (via commissioned surveys and even simple surveys) and
internal resources were favoured in the context of interactions with the authorities
(working groups, presenting argumentation material, parliamentary hearings).
Two characteristics emerge irrespective of the sector considered (small business,
craft, small- and medium-sized enterprises). First, the actors are unable to define
what falls within the ad hoc analysis of public policies and lobbying; the studies
undertaken and transmitted to elected officials in favour of the interests defended
are used to support dissemination.
The second characteristic is linked to the state’s role as a partner of so-called
‘independent’ expertise structures. One of the leading think tanks on ‘independent
economic analysis’, Coe-rexecode, emerged in 2006 following the merger of
two 50-year-old institutions. Of its funding, 15 per cent was provided by the
Chamber of Commerce and Industry of Paris Île-de-France (Chambre de commerce
et d’industrie de Paris Île-de-France [CCI Paris Île-de-France]), from which it had
partially originated. This public institution was under the authority of the state
and was administered by elected business leaders. Established as a partnership
between the professional organisation of the food trade and the trade unions
within the food industry in 2004, the occupations and skills observatory signed
a contract to provide prospective studies with the state in 2009. Similarly, in the
mid-1990s, the Higher Institute of Trades developed a craft innovation centre
234
managed on behalf of the Ministry on which it partially depended (the Institute

was established in partnership with the Ministry of Crafts, UPA and APCMA).
It is thus worth noting that, as in the latter case, the emergence of these ‘back
rooms’ was less associated with the desire to use new policy analysis tools than
with the will to participate in the legitimisation of the ongoing shaping of the
representation of the sector (Milet, 2008).
A state dependent on the expertise of business circles?
The literature on the effects of information resources in lobbying (Bernhagen

and Bräuninger, 2005) has shown the highly varied nature of the status accorded
to policy analysis across different structures. The authors postulate that the use
of expertise differs according to two constraints: the level of technical information
available for policy and for business groups on the one hand, and, on the other,
the reputation of the interest group vis-à-vis government, and of the ruling group
vis-à-vis its electorate. When a government has sufficient expertise resources,
economic interest groups favour self-restraint to avoid losing their reputation
by disseminating erroneous analyses. Existing studies suggest that, in the same
country, the degree of dependence of government in terms of expertise varies
across policy sectors, but also across the different decision-making structures.
Using ‘policy analysis’ rather than ‘lobbying’ provides new arguments that enable
a breaking away from the idea of a national model of expertise, and offers an
interesting avenue for overcoming the harmful effects of analyses in terms of
influence. Evaluation of the timing and degree of involvement of interest groups
in the generated expertise is independent of the quest for knowledge on the
potential effects on decisions.
The transformations of two specific public policies perfectly illustrate this
in the French case (Milet, 2013). The first began in the summer of 2007. The
president commissioned a former minister to prepare a report on the conditions
for implementing a ‘Small Business Act à la française’ expected to promote
SME access to government, then European contracts. The presidential project
of a Small Business Act (SBA) for Europe, established by Bercy following the
recommendations of a task force, was based on a typically French perception
of state intervention which failed to meet the expectations of the European
Commission and a majority of European partners eager to preserve the logic of
free trade. Subsequently, the initial French project had to be rapidly restructured.
It was only when the mechanisms were reorganised that European partners and
French representative organisations (UPA, CGPME) rallied around the mechanism
which was adopted in the fall of 2008. In terms of the decision-making process,
the SBA, which established a set of legally non-binding measures via ‘soft law’,
was thus hardly in line with the initial objectives of the French presidency,
despite the fact that the representative organisations had merely acted within the
framework of a reactive action.
235
The second policy is rather different: this is the promotion of reduced VAT rates
in the building renovation sector which was promoted by the main representative
organisation, CAPEB, from the mid-1990s. The expertise produced by the
organisation (via commissioned surveys and information gathered internally)
sought to show the government that tax revenue losses could not only be partially
offset by the effects associated with the economic stimulation of the sector, but
could also encourage employment. The analysis generated took a comparative
approach (with Luxembourg in particular). In both cases, however, the mechanism
adopted cannot be directly attributed, nor is it reducible, to the actions of
economic actors. Similarly, within a given sector, the status of expertise differs
depending on whether the analysis produced by the organisations drive public
policy (proactive action), or whether they are simply a response to a proposed
reform (reactive analysis).
The metamorphosis of economic interest groups

Deregulation policies, state reforms and the institutionalisation of French market
lobbying have all been involved in the metamorphosis of economic interest groups.
At the heart of this metamorphosis, the space occupied in the state configuration
presented at the beginning of this chapter has ceased to exist. Consultation
and forecasting systems have been dismantled, putting an end to governmental
dependency on data produced by economic interest groups. Questioned on this
point, employers are no longer aware of this past, nor do they feel that INSEE
can provide them with relevant information. The government, for its part, now
increasingly uses consultants to develop this data (Henry and Pierru, 2012). As
well as integrating consultants, the models used by economic interest groups are
proving to be increasingly fragmented, distanced from the past, and in the very
process of metamorphosis.
Dissemination of the evaluation model
In the late 2000s, European Association of Craft, Small- and Medium-Sized

Enterprises (Union Européenne de l’Artisanat et des Petites et Moyennes Entreprises
[UEAPME]) – the Eurogroup to which the most representative organisations
of craftsmen and SMEs of the member states adhered – embarked on the
implementation of an internal evaluation of European policies. Faced with a
deepening economic crisis, the predominant idea was that the arguments defended
by the group could be better taken into account if they relied on ‘statistics,
surveys and quantitative data’. The establishment of a ‘study unit’ was perceived
as a response to the dual competition in the statistical arguments production
market that emerged from the DG Enterprise of the European Commission and
databases already created by Eurochambres, BusinessEurope, and other economic
interest groups. However, even as the Eurogroup invested in opinion surveys
among entrepreneurs, the main French member organisations, that is, CGPME
236
and UPA, launched similar partnerships with external consultants at the national
level. The Europeanisation of the assessment mechanisms of the economic health
of different sectors primarily involved the use of ‘barometers’ rather than the
widespread use of statistical tools: the quantifiable data produced hardly relied
on objective data, but was the result of combined ‘staff judgements’ collected
from respondents’ perceptions when data on the constitution of representative
panels of craft enterprises and SMEs was collected. As such, assessment of the
effects of public policy and the economy was primarily based on opinion surveys.
The Europeanisation of policy analysis was also reflected in the dissemination of
national surveys from other member organisations of the Eurogroup. In the early
2000s, impact studies were commissioned by the French professional organisation
to challenge the studies undertaken by the European Commission. Managerial
techniques were first taken up by government officials with the European
Commission’s adoption of ‘new public management’ techniques (Sanchez Salgado,
2007); they were then adopted by interest groups.
At national level, economic interest groups began to use the same assessment
approach as public authorities, reflecting the transposition of policy categories
into collective action. While this process already existed, changes applied to the
constitution on 23 July 2008 were particularly reflected by the obligation for
ordinary legislation to develop impact assessments (Organic Law of 15 April 2009).
The economic actors endowed with sufficient resources seized this opportunity
to develop a ‘lobbying assessment’ within their repertoire of action. This original
assessment independently established government authorities through self-referral.
Unlike ‘technocratic’ and ‘participative’ assessments (Hassenteufel, 2008, 235), it
did not seek to improve and/or control policy but rather to influence law-making,
depending on what was promoted. Thus, the use of ‘anti-impact assessments’
grew, such as those established by organisations representing regulated professions
within the framework of the assessment of the growth and activity act (known
as the ‘Macron Act’) in 2015.
Changes observed in ‘applied analysis’
In terms of social interactions, the 2000s marked a turning point that saw France
shift from a system in which interactions between employers and trade unions were
viewed as a ‘by-product of conflict’ (in the words of D. Labbé) to a system of social
dialogue promoted by the state. This institutionalisation of forms of social dialogue
led to changes in the expertise provided by the employer organisations which
now provided the foundation, in part, for the production of labour legislation.
Inspired by European social dialogue mechanisms, the Larcher law of 31 January
2007 established a mandatory procedure requiring social partners to consult
with the government before making laws. In the fields of vocational training,
employment and working conditions, the government provided a guide (with a
‘diagnosis’ and the objectives and main options selected) to the social partners.
These partners could decide whether or not to initiate negotiations in order
237
to reach an agreement transposable into the legislation. Once again, however,

the established regime remained somewhat ambivalent. Indeed, the restoration
of social dialogue shifted the centre of gravity of the defended positions from
parliament to social partners. Moreover, employee unions deplored the fact that
the agreements negotiated were directly drawn from the proposals developed
and initiated by employer organisations. When assessing draft laws, employers’
‘common position’ was the main working text used. The ambivalent status of
the state arose from the fact that it could use the right of amendments or the
declaration of state emergencies to circumvent mechanisms, and also because the
new regime was not enshrined in the constitution.
Besides these constitutional and legislative developments, the analysis developed
by economic interest groups was also affected by major changes in the political
field. Two of these stand out: recognition of different forms of participatory
democracy and the institutionalisation of the electoral calendar in which
presidential and legislative elections were synchronised. One policy provides a
perfect illustration of the impact of participatory democracy on interest group
activities. Energy policies – which revolve around the role of nuclear power in
France – have increasingly resorted to debate to enable representatives of the
interests concerned to express themselves. The law on energy transition thus
contains mechanisms through which MEDEF and pro-nuclear representatives
are overrepresented. These new arenas provide numerous forums through which
the incentives and different scenarios of proposed energy policies can be assessed
and objectified.
The recent electoral calendar further reinforces the need for a forward-looking
approach. Economic interest groups have used electoral conditions to present
their ‘projects’ over a five-year period to prospective candidates (Courty and
Gervais, 2016). These projects are as much a review of the previous legislature
– the state of positive law, identification of abandoned texts, formulation of the
necessary requirements for the sector – as well as a projection of the five-year
term about to begin. At the heart of all these documents, some of which are
presented as ‘programmes’, is another facet of policy change. Economic interest
groups also view public policies as multi-pronged, comprising economic, legal,
social, environmental and tax issues; subsequently, it has become increasingly
difficult for a single law or a single department to address these issues. Through
their development, policies help revive parliament’s power.
Conclusion
Following dramatic changes in the twenty-first century, the French political
field has done away with the last traces of the role played by economic interest
groups within the state. While these groups no longer intervene upstream public
policies as the accredited suppliers of socio-economic data necessary for the
formulation of public policy, they have nevertheless supported the evolution
of the political system of the Fifth Republic by co-producing policies owing
238
to their maintained claim of expertise in their respective fields, and their use
of evaluation and simulation methods. There has been no specific need for the
creation of specialised structures to address these activities; on the contrary, they
have extended and multiplied across directions, services and consultancy services,
depending on the skills required.
Notes
1 The gender perspective about interest groups remains underdeveloped even if organisations
representing women entrepreneurs exist (Femmes chefs d’entreprises [FCE France]). A limited
number of studies show the over-representation of men at the head of these organisations.
Others provide some information about attempts at feminising specific jobs or discuss the role
of spouses in economic activities, especially the craft sector.
2 In this sense, interprofessional organisations have thus communicated to their members to
provide expertise on changes in the rules of equality between men and women (see Medef ’s
practical guide which proposes ‘a general outline for the implementation of the new Equal
Opportunities’, June 2012, and the leaflet ‘UPA equality between women and men in local
businesses: A reality and an opportunity’, June 2014.
3 The UPA was transformed in November 2016 to U2P, following the accession of the UNAPL
(liberal professions).
References
Benamouzig, D. (2012) ‘Du grand soir au clair-obscur: Expertise économique
et privatisation bureaucratique de l’assurance maladie’, Actes de la recherche en
sciences sociales 193–6, 56–73.
Bernhagen, P., Bräuninger, T. (2005) ‘Structural power and public policy: A
signaling model of business lobbying in democratic capitalism’, Political Studies
53, 43–64.
Courty, G. (2006) Les Groupes d’intérêt, Paris: Coll. Repères, La Découverte.
Courty, G., Gervais, J. (2016), ‘Les représentants d’intérêt et la campagne
présidentielle de 2012: Rapports au politique et formes de coopération avec
les candidats’, Politix 113(29), 117–39.
Courty, G., Suleiman, E. (1997) L’Age d’or de l’État: Une métamorphose annoncée,
Paris: Coll. L’histoire immediate, Le Seuil.
Dubos, J. (2002) ‘La Fédération des Industriels et des commerçants français entre
groupe de pression et syndicat d’union patronale’, in J. Garrigues (ed) Les groupes
de pression dans la vie politique contemporaine, en France et aux Etats Unis, de 1820
à nos jours, Rennes: Presses universitaires de Rennes.
Duclos, L., Mériaux, O. (2001) ‘Autonomie contractuelle et démocratie sociale:
Les implicites de la “refondation”’, Regards sur l’actualité 267(Jan), 19–34.
Greenan, N., Walkowiak, E. (2010) ‘Les structures organisationnelles bousculées
par les nouvelles pratiques de management?’, Réseaux 4(162), 73–100.
Grossman, E., Saurugger, S. (2004) ‘Les groupes d’intérêt français: Entre exception
française, l’Europe et le monde’, Revue internationale de politique comparée 4(11),
507–29.
239
Guillaume, S. (2004) Le petit et moyen patronat dans la nation française de Pinay à

Raffarin, 1944-2004, Paris: Presses Universitaires de Bordeaux.
Hassenteufel, P. (2008) Sociologie politique: L’action publique, Paris: Armand Colin.
Henry, O., Pierru, F. (2012) ‘Les consultants et la réforme des services publics’,
Jobert, B., Muller, P. (1987) L’Etat en action: Politiques publiques et corporatisme,
Keeler, J.T. (1987) The Politics of Neocoporatisme in France: Farmers, the State and
Agricultural Policy-making in the Fifth Republic, New York: Oxford University Press.
Lefranc, G. (1976) Les Organisations patronales en France: du passé au présent, Paris:
Payot.
Lowery, D. (2013) ‘Lobbying influence: Meaning, measurement and missing’,
Interest groups and advocacy 2(1), 1–26.
Margairaz, M. (1991) L’État, les finances et l’économie: Histoire d’une conversion
1932–1952, Paris: Comité pour l’Histoire économique et financière de la
France, Vols 1–2.
Michel, H. (2006) La cause des propriétaires Etat et propriété en France, fin XIXe-XXe
siècle, Paris: Belin.
Milet, M. (2008) ‘Parler d’une seule voix: La naissance de l’UPA et la (re)
structuration du syndicalisme artisanal au tournant des années 1970’, Revue
Milet, M. (2013) ‘La configuration de la décision. Retour sur la question de
l’influence des lobbies à partir des politiques européennes à l’égard des PME’, in
Mélanges en l’honneur du professeur Jacques Chevallier: Penser la science administrative
dans la post-modernité, pp. 167–77, Paris: Librairie Générale de Droit et de
Jurisprudence (LGDJ).
Muller, P. (1982) ‘Comment les idées deviennent-elles politiques? La naissance
d’une nouvelle idéologie paysanne en France, 1945–1965’, Revue française de
science politique 32(1), 90–108.
Offerlé, M. (1998) Sociologie des groupes d’intérêt, Paris: Montchrestien (Clefs).
Offerlé, M. (ed) (2011) L’espace patronal français: Acteurs, organisations, territoires,
Paris: Ministère du travail, DARES.
Offerlé, M. (2013) Les Patrons des patrons: Histoire du Medef, Paris: Odile Jacob.
Palier, B. (2002) Gouverner la sécurité sociale: Les réformes du système français de
protection sociale depuis 1945, Paris: Quadrige, Presses universitaires de France.
Restier-Melleray, C. (1990) ’Experts et expertise scientifique: Le cas de la France’,
Robert, C. (2008) ‘Expertise et action publique’, in O. Boraz, V. Guiraudon
(eds) Politiques publiques 1: La France dans la gouvernance européenne, pp. 309–35,
Salisbury, R. (1969) ‘An exchange theory of interest groups’, Midwest Journal of
Political Science 13(1), 1–32.
240
Salisbury, R.H. (1991) ‘Putting interest back into interest groups’, in J. Cigler
Allan, A. Loomis Burdett (eds) Interest Group Politics, Washington: Congressional
Quarterly Press.
Sanchez-Salgado R. (2007) Comment l’Europe construit la société civile, Paris: Dalloz.
Saurugger, S. (2002) ‘L’expertise: Un mode de participation des groupes d’intérêt
au processus décisionnel communautaire’, Revue française de science politique 52(4),
375–401.
Schmidt V. (1996) From State to Market? The Transformation of French Business and
Government, New York and London: Cambridge University Press.
Schmitter, P.C., Streeck, W. (eds) (1985) Private Interest Government: Beyond Market
and State, London: Sage.
Suleiman, E. (1979) Les élites en France: Grands corps et grandes écoles, Paris: Seuil.
Volle, M.l. (1982) Histoire de la statistique industrielle, Paris: Economica.
Wilson, F.L. (1987) Interest Group Politics in France, Cambridge: Cambridge
University Press.
Woll, C. (2006) ‘La réforme du Medef: Chronique des difficultés de l’action
collective patronale’, Revue française de science politique 56(2), 255–79.
Zalio, P.-P. (2004) ‘D’impossibles notables? Les grandes familles de Marseille face
à la politique 1860–1970’, Politix 17(65), 93–118.
Zittoun, P. (2001) ‘Partis politiques et politiques du logement, échange de
(October), 683–706.
241
FOURTEEN
NGOs, civil society and policy analysis: from

mutual disinterest to reciprocal investment
Laurie Boussaguet and Charlotte Halpern
Introduction
In this chapter, we address the role played by non-governmental (NGOs) and civil
society organisations (CSOs) in the development of policy analysis in France. A
brief overview of the existing literature about the French context suggests that
these organisations have an ambivalent relationship to the policy process. Scholarly
work paid little attention to these actors until relatively recently. Two major
reasons were repeatedly brought forward in order to account for such mutual
disinterest (Callon et al, 2013). First, forms of decision-making and state–society
relationships in the context of the Fifth Republic offered little opportunities
for non-state actors to shape policy-making. CSOs and NGOs were primarily
considered as service providers at policy implementation stage, by contrast with
the formal role granted to workers’ unions and professional organisations in the
policy process after the Second World War. Second, the French neo-corporatist
model of policy-making favoured the participation of a limited number of
organised CSOs to policy formulation (Hayward and Watson 1975), whereas
the vast majority of CSOs and NGOs occupied marginal positions in the policy
process and preferred other forms of interests’ representation, such as protest
(Wilson, 1983). While the former were often referred to in scholarly literature
as policy insiders and were, as such, the focus of much attention in scholarly
work about the functioning of the state and policy-making in France, the latter
were considered as policy outsiders and only included in policy studies insofar
as they acted as veto-players or as agents of policy change.1 Decentralisation
reforms somewhat confirmed this division of tasks. Following the 1982 laws, the
generalisation of ‘public policy by delegation’ (Lorrain, 2005) strengthened the
critical role of non-state actors – both private and non-profit organisations – in
the development of policy offer and the provision of public services at the local
level (see also Gaudin, 2007). Yet the formulation of public policies themselves,
and the production of policy analysis, remained concentrated at the national level
and in the hands of higher civil servants – generalists and/or specialists – and
professionals (Douillet et al, 2012). As a result, until recently, the relationship
between NGOs and policy analysis was characterised either by a mutual lack of
interest or by strong distrust.
243
Recent work done on CSOs and NGOs suggests, however, their new and
growing interest in the production of knowledge and expertise, as well as for
their formal inclusion into state-led consultation processes and organisations across
levels of government (Hély, 2009; Laville and Salmon, 2015). The strategic use of
policy analysis also proves particularly instrumental in the competition between
organisations for public subsidies and funding. This progressive shift has been
observed across a large variety of policy domains, and also concerns organisations
that explicitly refused to cooperate with state organisations in the past. It increased
over the recent period in a context of reduced available public subsidies and the
post-2008 crisis. In the absence of any systematic work on these groups’ uses
of policy analysis across policy sectors, this process remains undocumented: for
those CSOs and NGOs that are involved in the production of policy analysis,
what does it entail? What are their uses of policy analysis, and how is it included
in their action repertoires?
This chapter seeks to explain the growing interest of NGOs and CSOs for
policy analysis as a legitimate form of action repertoire, and it return, how it
was analysed in policy studies. The first section discusses the origins and later
development of the voluntary sector in the French context. We then proceed
to examining the role of CSOs and NGOs in the production of policy analysis
until the early 1990s. In the third section, we explore the shift that took place
recently by looking in more detail into these groups’ effective contribution to
policy analysis and how it was analysed in policy studies.
CSOs and NGOs in the French context: origins and latest

developments
Beyond the French context, it is widely agreed that CSOs and NGOs share some
common features, in the sense that they are voluntary associations, independent
and not-for-profit (Saurugger, 2007; Kohler-Koch, 2008). Trade unions, political
parties and organisations representing specific professional interests are also
included in this broad understanding. Notwithstanding these generic approaches
to CSOs and NGOs, important variations can be found from one political system
to another, as their origins are deeply rooted in long-term historical and political
processes that constitute state–society relationships in a given polity. In the French
context, this is very much linked to the development of the voluntary sector in
the context of the 1901 Law on the Freedom of association.
The invention of the voluntary sector
The emergence of the voluntary sector in the French context is closely related to
post-revolutionary debates about individual and collective rights, and to forms of
interests’ representation beyond the state’s realm (Hayward, 2007). Following the
1791 Le Chapelier Law, any pre-existing organisations (religious, professional,
political and so on) were abolished and a de facto ban on the right to strike and
244
NGOs, civil society and policy analysis
to form an association was introduced. Until 1884 (Waldeck-Rousseau Law),

strikes, demonstrations and associations were considered as a major threat to
successive political regimes, thus explaining why various groups could only be
incorporated in policy-making on a limited basis. Indeed, the voluntary sector
emerged at the intersection between, on the one hand, a strong distrust in the
ability of self-interested factions to generate desirable policies in the general interest
and on the other hand, increased demands stemming from an active citizenry in
favour of additional collective rights (Hazareesingh, 1994; Rosanvallon, 2004).
The 1901 Law on Freedom of Association (the so-called ‘Association Loi 1901’)
was introduced in order to allow for the development of community-based
organisations while at the same time accommodating the legacy of the French
Revolution. The Association Loi 1901 opened new opportunities for society-based
organisations, provided that they complied with specific procedures, including
clearly defined statutes. Since then, a large variety of organisations (trade unions,
political parties, interest groups and citizens’ initiatives and so on) developed as
part of the 1901 Law umbrella that still constitutes, to this date and, with the
exception of religious associations,2 a cornerstone of state–society relationships
in the French context. This broad category was refined through successive pieces
of legislation, thus leading to multiplying the number of subcategories that are
defined according to the rights (for example, fiscal or judicial) granted by the
state and other public authorities at various levels of government.
Insofar as they contribute to explaining variations in the rights and obligations
these organisations enjoy, some differences are worth mentioning. First, although
under the Association Loi 1901 organisations are under no obligation to formally
register with public authorities, only registered organisations enjoy full legal
capacity and as such, are allowed to develop relationships with third parties, being
public or private. In addition, only a small number of registered organisations are
recognised as being of public utility (some 2,000 in 2015), which grants them
access to extended legal capacity and rights, including the ability to manage
state property and funding for general and public interest activities. Second,
some additional variations are observed according to these organisations’ main
focus (for example, sports and youth, health and social care, nature protection,
women’s rights and consumers), their scale of intervention (for example, local
versus regional/national), and intra-sectoral levels of cooperation (for example,
presence or absence of umbrella organisations). As of late, citizens’ associations
emerged as a specific type of community-based organisation that pursues an
objective that goes beyond the satisfaction of its members and aims at contributing
to the common good.
Notwithstanding this large variety of CSOs and NGOs, this chapter adopts a
restrictive approach and leaves out trade unions, interest groups, as well as political
parties and think tanks, each of those being addressed by other contributors to
this volume.
245
The voluntary sector in France: recent trends
A brief overview of the voluntary sector today gives some additional insights to
the above-mentioned changes. It confirms a high level of diversity across and
within policy sectors, as well as recent rapid and profound restructuring processes.
This broad overview also suggests the critical role played by the Association Loi
1901 status in shaping relationships with public authorities.
According to the latest census done by the National Institute of Statistics and
Economic Studies (INSEE, 2010), the voluntary sector includes a large number
of organisations, which amounts to some 1.3 million in 2012 and, together, add
up to 21 million members and some 11 million volunteers (see also CPCA, 2012;
Bazin and Malet, 2015). Additional differences can be observed when considering
sources of revenues (for example, membership fees, donations, public grants),
the amount of revenues, and their thematic scope and scale of intervention (for
example, from the neighbourhood to the national scale) (Tchernonog, 2012).
Historically, the vast majority (66 per cent) of organisations intervene in the
field of culture, sports and leisure, while only a minority seeks to promote
collective rights or professional interests, or to provide humanitarian, charity
and healthcare services. The former are primarily organised at neighbourhood
level, they experience a rapid turnover rate, manage low levels of income and
are highly dependent on public grants, while the latter are primarily organised at
national level, benefit from a larger variety of income sources and manage high
levels of income. Indeed, some 22 per cent of existing associations manage a
yearly income of less than €1,000 that is, 0.2 per cent of the total budget of the
voluntary sector, while 2 per cent of existing associations, which are primarily
active in the field of health and social care manage a yearly income of more than
€500,000, which is more than 70 per cent of the total budget of the voluntary
sector. In 2012, the voluntary sector added up to a total revenue of €85 billion,
which relied primarily on the provision of services (60 per cent), public grants
(25 per cent), membership fees (10 per cent) and donations (1.9 per cent).
By contrast with other western democracies, corporate sponsorship is little
developed in the French context. It only benefits some 9 per cent of the biggest
organisations, and more specifically national-based organisations from the
humanitarian sector. Yet, when looking more specifically at public (tendering and
grants) versus private (membership fees, donations and users) sources of income,
the dependency to public authorities in terms of income sources remains above
the 50 per cent threshold. Over the past decade, and as a result of declining (or
less stable) membership income on the one hand, and of decreasing income from
the public sector, the voluntary sector tends to increase private revenue sources by
increasing service rates. Among those public authorities that directly contribute
to funding the voluntary sector, the state’s contribution declined from 15 per cent
of total income sources in 1999 to 11 per cent in 2012, and that of municipalities
from 15 per cent in 1999 to 11.5 per cent in 2012. By contrast, contributions
of the counties (départements) have known a constant increase over the same
246
period, as a result of their increasing competences in the social and the health
policy domains. While the amount of grants from public authorities is constantly
decreasing, their support to the voluntary sector increasingly takes the form of
public tendering or delegation in exchange for the provision of services. This is
particularly the case for those groups who are declared to be of public interest.
This broad overview of recent changes taking place within the voluntary sector
suggests a shift in the role and function of CSOs and NGOs in the policy process
(Laville and Salmon, 2015). Indeed, grassroots organisations increasingly tend
to operate as policy insiders, especially at implementation stage and, to some
extent, during policy formation and evaluation. This is primarily explained in
a context of reduced public funding capacity across levels of government, and
the need for CSOs and NGOs to demonstrate value for money. It also suggests
increased inequalities within and across policy domains, in terms of resource
distribution and levels of professionalisation. Yet it provides little information
regarding other action repertoires (protest, judicialisation) and contradicts the
view that is commonly found in the academic literature regarding these groups’
marginal role in the policy process.
The high level of variation observed between policy domains constituted a major
obstacle to cross-sectoral comparative studies. As a result, we draw on the existing
academic and practitioners’ literature in order to illustrate those few trends that
are widely shared across time and policy domains while at the same time seeking
to highlight profound differences through the selection of most-different cases.
Civil society organisations and policy analysis: a long period of

mutual lack of interest
In this section, we seek to explain why, until a recent period, the relationship
between CSOs and NGOs and policy analysis was characterised by mutual lack
of interest. We argue it is due, one the one hand, to policy analysis itself, and
the lack of interest of practitioners and academics vis-à-vis these organisations,
and on the other hand, to CSOs and NGOs themselves as policy analysis was
often disregarded as a possible – or even desirable – tool in developing their own
strategies.
The blindness of emerging policy studies for CSOs and NGOs
The blindness demonstrated by policy studies for the role of CSOs and NGOs
goes back to the discipline’s origins in France. Following its formal introduction
in 1981 (Boussaguet and Surel, 2015), policy studies followed a slightly different
path in the French context to that of its American counterpart. It was immediately
considered an academic field of study that was to be strictly demarcated from
policy analysis itself, including from the pre-existing tradition of public and
rational policy analysis within the Plan Commission (Commissariat Général au
Plan).3 Furthermore, its import into the French social sciences academic context
247
led to some adjustments in order to accommodate existing intellectual traditions,

such as organisational sociology, Marxism or administrative science (Leca and
Muller, 2008).
This had some long-term impact over policy studies themselves,4 such as
the rejection of rational choice approaches, a stronger interest in actors and a
preference for a top-down perspective on policy-making, which by extension
justified the marginalisation of CSOs and NGOs. First, the policy process was
primarily considered from a ‘statist’ perspective, in which civil servants, and to a
lesser extent elected representatives, plaid a prominent role. This somewhat echoes
practitioners’ views as well as their lack of faith in the ability of self-interested
groups to generate desirable policies in the general interest. Second, policy studies
in the French context reproduced, to some extent, the shortcomings of the ‘French
model of policy-making’ insofar as it primarily focused on a small number of
actors. In this perspective, that was very much influenced by the work of Jobert
and Muller (1987), policy-making was characterised by the centrality of the state,
the critical role played by corporatist arrangements, and a strong level of sectoral
autonomy in the choice and selection of policy priorities and solutions. In other
words, policy-making results from the close cooperation between a ruling state
elite and organised interests.
The notion of ‘circles’, as developed by Muller (2009), best exemplifies this
hierarchical perspective of the policy process. The first circle includes key
decision-makers (for example, members of the executive and their respective
staff). They are considered as the main instigators of major political decisions and
critical policy choices and as prominent actors in setting policy priorities and
selecting preferred modes of action. The second circle is constituted by sectoral
administrations (education, agriculture, environment, industry, defence and so on),
in which higher civil servants – both generalists and specialists – play a prominent
role in order to adjust sectoral interests to the wider political requirements and
policy priorities. Non-state actors are included in the analysis as part of the third
circle,5 yet this notion refers to organised interests in the framework of corporatist
arrangements, that is, professional organisations, trade unions or experts, insofar
as they are identified as major veto-players or drivers of policy change, whereas
the vast majority of CSOs and NGOs are considered as policy outsiders.6 In his
account of the changes taking place in the agricultural policy domain in the late
1950s, Muller examined in detail the growing role of the Catholic Agricultural
Youth Organisation (Jeunes Agriculteurs Catholiques ) in developing an alternative
policy programme that was grounded in a changed approach to the farmers’ role,
namely that of a business manager as opposed to a householder, and how this
CSO successfully championed a major policy reform (Muller, 1984).
Similar studies were undertaken in these early stages of French policy studies in
order to explore the relationship between other professional organisations (civil
engineers, doctors, teachers and so on) and policy developments across sectors.
Interestingly, the ability of these groups to produce information and knowledge
about their respective domains and to develop alternative policy programmes is
248
identified as a resource conducive to success, insofar as it is strategically mobilised

jointly with other resources (such as protest) as part of influence-seeking strategies.
By contrast, little attention is devoted to non-professional organisations and by
extension, to those transversal policy areas such as environmental, social and
cultural policy areas, in which citizen-led organisations played a critical role in
early policy developments.
Civil society organisations far from policy analysis
This lack of interest by CSOs and NGOs in French policy studies resonates
with that of a vast majority of CSOs and NGOs in policy analysis as a resource
commonly used in their action repertoires. In her extensive work on the
Associations Loi 1901, Barthélémy (2000) convincingly argues that only a minority
of these groups are ‘likely to take, at least partially, a defence function, either
“participatory” (within the decision-making circuits) or “tribune” (without
willingness to invest power spheres). The assimilation of associations to forms
of social experimentation and identity or militant expression proceeds from
a largely illusory vision’. Policy studies in and about France echoes this view.
These groups are considered as ‘policy outsiders’ in the policy-making process,
which is primarily shaped by internal dynamics in combination with political
and institutional features. In this respect, comparative political research often
highlighted the so-called ‘French exceptionalism’, that is a tendency of the
‘contentious French’ (Tilly, 1986) being more prone to unconventional forms of
participation and the choice of protestation as a preferred and dominant action
repertoire. It seems so true that some authors suggested adding a third model
to the two already existing ones (‘pluralist’ and ‘neo- corporatist’) in order to
better qualify state–society relations in the French context, namely the ‘protest
model’ (Wilson, 1983). In this perspective, organised groups primarily sought
to indirectly exert pressure by mobilising public opinion through the extensive
use of protest, including demonstrations, strikes, sit-ins and boycotts. Unlike
the situation observed elsewhere (Nelkin and Pollack, 1981), they rarely used
expertise and knowledge production as part of their influence-seeking strategies.
According to these authors, two main factors accounted for this overrepresentation
of protest in the French case. First in the context of the Fifth Republic, the
French state constitutes an archetypically strong, centralised and unified political
opportunity structure. Insofar as few venues are opened to alternative expertise
and policy challengers, technocratic forms of decision-making encourages the
use of disruptive strategies as seen as the last resource of the powerless (Kitschelt,
1986). Second, due to internal power struggles, organisations themselves choose
to remain policy outsiders and prefer politicisation as an alternative form of
influence-seeking strategy. Examples from two contrasting policy domains –
environment and culture – offer a good opportunity to further examine these
organisations’ ambivalent approach to policy analysis.
249
CSOs and NGOs in the environmental sector are particularly representative

of the so-called ‘protest model’ (Hayes, 2002; Filleule, 2002). Also, due to the
transversal dimension of environmental issues, it offers a good opportunity to
understand the role of policy analysis in the strategies developed by CSOs and
NGOs in order to challenge forms of policy-making in those policy sectors
that are often considered as paradigmatic cases of the French public policy
model (for example, transport, energy or agriculture) (see also Szarka, 2001).
The development of the French ecologist movement is deeply rooted in a vivid
network of grassroots organisations structures, which, together, sought to ensure
some level of coordination at national level under the leadership of a single
umbrella organisation, that is, France Nature Environment (FNE)7 (Ollitrault,
2008). As such, these organisations played a pivotal role in the creation of the
Ecology Ministry’s central and regional services (Lascoumes and Le Bourhis, 1997;
Charvolin, 2003), so much so that this administration was often referred to as
an ‘activist administration’ (Spannou, 1991). In this policy’s early developments,
grassroots’ organisations provided the Ministry with two critical resources: on the
one hand, detailed information on the state of the environment and if and how
policies were being implemented, and on the other hand, power to denounce
non-compliance through protest and litigation (Lascoumes, 1994). As non-
profit organisations (Associations Loi 1901), these organisations could apply for
public funding and formal recognition by public authorities (agrément) as a way
to gain access to the judiciary system. Some of them successfully engaged in the
procedure by which they were recognised to be of public utility, and through
which they could access additional legal rights. As such, they also contributed
to expanding the policy offer and to delivering specific services in cooperation
with local authorities.
Environmental CSOs ad NGOs, however, are usually considered in policy
studies as watchdogs rather than service providers. In some areas such as the
management of natural resources (water, fauna and flora and so on), they were
marginalised as they competed with specialised fields of knowledge and expertise,
some of which were trained in Grandes Écoles. Also, a majority of ecologist activists
and organisations in France contested FNE’s strategy and highlighted the need
to maintain their position as outsiders to mainstream political parties and the
institutional system (Halpern, 2016). By contrast with FNE, a new generation
of environmental organisations (for example, the French branch of Friends of
the Earth (FoE) and Greenpeace) was less interested in the preservation of the
state of nature but sought for increased democratic participation and alternative
sources of expertise in policy-making as part of an anti-nuclear protest. They
favoured protest and litigation over policy analysis, and publicly opposed FNE’s
attempts to cooperate with state authorities.
By contrast, the cultural policy domain shows why CSOs and NGOs developed
a growing interest in policy analysis as part of their direct involvement in policy
implementation. In this policy domain, the high number of associations is strongly
related to this policy’s origins and initial goal which is the principle of cultural
250
democratisation.8 Following the mobilisation of extraparliamentary opposition

during the 1960s, alternative forms of cultural policy-making and implementation
were favoured in order to operationalise the idea of ‘access to culture for all’.
Successive generations of cultural activists actively engaged in the creation of
large networks across cultural policy subfields (theatre, classical music and so on).
In some cases, this was done in cooperation with formal cultural institutions,
but not systematically. Indeed, the creation of CSOs was strongly encouraged
in order to achieve increased spatial coverage throughout the national territory
and to explore a greater diversity of forms of cultural expression, including those
stemming from civil society and citizen participation (Arnaud et al, 2015).9
Yet such support did not prevent the permanence of strong hierarchies and
the creation of new inequalities between CSOs (Dubois, 2015), which are also
– but not exclusively10 – shaped by relationships with state authorities and the
use of policy analysis as an influence-seeking strategy over the selection of policy
priorities and tools. Only a small minority of CSOs did develop policy analysis,
while a vast majority focused on policy offer, including the provision of specific
services and the management of cultural facilities on behalf of public authorities.
Furthermore, and with the exception of a few figures, policy formulation remains
in the hands of higher civil servants, some of whom were trained as generalists
at the ENA, others stemmed from specialised schools (École du Louvre, École
normale supérieure, École des Chartes and so on) and some were competitively
selected. In addition, local state authorities (Préfets) and, following the 1982
decentralisation laws, public authorities across levels of government, often made
a strategic use of the distribution of public subsidies and of contractualisation
procedures. This proved particularly instrumental in ensuring social control
(for example in the suburbs of large cities) (Gaudin, 2007) and in developing
clientelistic networks and sustaining political leadership (Pinson, 2009).11
Upon closer attention, these approaches tend to overestimate the constraining
role of political opportunity structures. Yet whether active as service providers or
as watchdogs, NGOs and CSOs remain largely secluded from policy formulation
and policy analysis is not developed as a preferred mode of action. In the following
section, we examine how CSOs and NGOs have progressively emerged as policy
producers in the French context.
The increasing use of policy analysis as an action repertoire

The growing interest of CSOs and NGOs for policy analysis occurred as a result
of a continued shift in state–society relations since the 1980s onwards. Following
the claim recently made by Laville and Salmon (2015), these organisations can
no longer be categorised as either service providers or as protesters, but directly
contribute to restructuring forms of policy-making as a result of their changed
role in the policy process. Several factors are often highlighted in order to explain
how these actors’ position as an alternative source of expertise and knowledge
strengthened over time in the French context. First, some CSOs and NGOs
251
have been able to influence policy-making by blurring the limit between ‘policy
insiders’ and ‘policy outsiders’. In the case of some women’s groups, the presence
of ‘femocrats’ within the state apparatus has been repeatedly highlighted in the
literature (McBride Stetson and Mazur, 1995; Bereni, 2009; Banaszak, 2010).
Second, a series of environmental health crisis (for example, bovine spongiform
encephalopathy, or ‘mad cow disease’) led to vivid debates about forms of decision-
making and to increasing demands in favour of more diverse forms of expertise
and knowledge in policy-making processes (Gilbert and Henry, 2009). Third,
European and International law on the right to have access to information fostered
the opening of the policy-making process at the domestic level. Notwithstanding
some variations across policy domains, CSOs and NGOs were granted a formal
role in the decision-making process, mostly as observers. The introduction of ad
hoc consultation processes and participatory devices offered additional channels
of access to the policy process.
Third, Europeanisation (Saurugger, 2007; Balme and Chabanet, 2008) and
devolution (Douillet et al, 2012) processes also increased opportunities for these
organisations to systematically cooperate with public authorities in order to
jointly implement public policies across policy domains. Even though only a
limited number of CSOs and NGOs effectively engage in policy formulation
activities, and were able to compete for leadership with higher civil servants for
knowledge production and policy analysis, a large proportion of CSOs and NGOs
now ensures the distribution of policy offer and service provision throughout
the territory.
The development of alternative action repertoires
All these factors were critical in fostering a shift in the NGOs and CSOs
preferred action repertoires across policy areas, including litigation and knowledge
production. Following the opening of additional judicial opportunities, CSOs
and NGOs developed a strategic use of law and litigation. As convincingly argued
by Israël (2009), this does not result from a mere legal translation of a political
struggle but it also has some long-term impact on forms of mobilisation in the
French context. While it still counts as a reactive strategy, the preparation of trials
did foster the accumulation and the codification of information and knowledge.
Expert testimony during trials was seen as an opportunity to produce collective
memory and to develop a new series of arguments that would allow for activists
to be considered as credible contributors in the judicial arena.12 Alliances with
lawyers and civil servants that supported public interest lawyering were sought
after in order to promote changes in the jurisprudence in various areas of law. In
the case of immigration law, for example, the creation of organisations such as the
Immigrant Information and Support Group (Groupe d’information et de soutien des
immigrés [GISTI]) led to the accumulation of specific cases in order to consider
the variety of legal issues to be overcome and to develop a strategic approach to
litigation (Israël, 2009).13
252
In addition to litigation, demands in favour of the opening of the policy-

making process directly contributed to strengthening the role of NGOs and CSOs
within the policy process while at the same time accelerating a shift from protest-
oriented towards more policy-oriented strategies. By systematically engaging in
the production of policy analysis in their respective area of expertise, CSOs and
NGOs highlighted its role as a resource for collective action (Mouchard, 2002).
More specifically, by organising conferences and producing information bulletins,
CSOs contributed to data gathering and knowledge production, to identifying best
practices and to commissioning new or alternative research which, together, added
up to credible expertise in the policy-making process. Once again, the example
of the women’s movements is representative of this broader claim. Although the
use of protest has long been considered a preferred action repertoire, notably by
the second wave of feminist activists, women’s movements increasingly chose to
develop knowledge production and expertise as part of women’s rights campaigns.
In the case of sexual violence, these groups played a pioneering role in producing
knowledge about this issue by drawing on the information received through
the hotline and the shelters they helped create (Boussaguet, 2009). Indeed, the
strategic use of knowledge proved critical in order to influence the policy agenda
and shape policy solutions. The support they received from formal arenas also
helped amplify and transmit feminist expertise and discourses to policymakers
(Boussaguet, 2013) – a trend that is also exemplified by the decisive role played
by the Women’s Labour Committee (Comité du travail féminin) in the case of the
equality at work policy (Revillard, 2009).14 In the field of health policy, the
production of policy analysis accelerated the recognition of CSOs and NGOs as
credible contributors within the arena of credential expertise about rare diseases
or about the formal recognition of afflictions resulting from exposure to various
sources of pollution (Huyard, 2009; Gilbert and Henry, 2009). AIDS activism
fostered lay participation in biomedical research and the reform of clinical trials
(Favre, 1992; Callon et al, 2013). The growing interest of NGOs and CSOs for
policy analysis directly contributed to these groups’ professionalisation, that is,
the accumulation and concentration of a distinctive knowledge, the development
and codification of specific statues and the growing criticism against amateurism
(Hély, 2009).15 This is particularly true in those policy domains in which CSOs
and NGOs already played a meaningful role during policy implementation. In
the medico-social field for example, they run three-quarters of private medical
and social accommodation (nursing homes, kindergartens, hospitals and so on),
a dense network of home care for the elderly and a near monopoly in care for
people with disabilities and in distress.
The growing interest for policy analysis has been observed across a large
variety of policy domains, and also concerns organisations that explicitly refused
to cooperate with state organisations in the past. It is increasingly considered as
a major resource as part of fundraising and monitoring activities, as well as for
those that are formally included in consultation processes and organisations. To
be sure, such evolution can also be analysed in a classic Weberian perspective,
253
as a progressive autonomisation process. What we see, however, is that CSOs

and NGOs remain more action-oriented than oriented towards the production
of knowledge, thus explaining the continued disconnect with policy analysis.
Whereas their role in policy implementation is firmly institutionalised across
policy sectors, their role as policy producers is more contested for reasons that
are both external – for example, veto actors within the politico-administrative
system – and internal – for example, debates within each organisation about their
role and function. The rapidly evolving role of umbrella organisations from a
mere platform wholly dependent upon its members towards a fully developed
organisation that seeks to develop resources through alternative influence-seeking
strategies, including policy analysis, best exemplifies this tension.
In the case of addictology for example, the growing interest for policy analysis
and the recognition of its strategic role as part of influence-seeking strategies led
to the major organisational changes and the strengthening of a single umbrella-
organisation, the Fédération Addiction (Gaubert, 2014). The main rationale for
merging three pre-existing organisations explicitly resulted from the need to
adjust to new financial constraints while at the same time developing a global
approach to addictions in order to overcome barriers between different sources
of professional expertise and achieve greater impact on the policy-making
process. The production of knowledge for policy-making, including policy
recommendations and influence-seeking strategies, captures a growing share
of the organisation’s resources with some profound impact on internal power
relations between the umbrella organisation and its members as well as on its
external image.
Also, the strengthening of umbrella organisations and increased attention
devoted to policy knowledge and production as part of CSOs and NGOs activities
is closely related to the opening of new opportunities in the policy-making
process, as exemplified in the environmental policy domain. Since the late 2000s,
debates about ecological democracy accelerated the institutionalisations of large
environmental NGOs (for example, FNE, Friends of the Earth and so on) as
‘policy insiders’ (Boy et al, 2012). The introduction of mandatory policy review
mechanisms and the strengthening of inquiry procedures at implementation stage
directly benefited those organisations that had been able to establish themselves as
legitimate policy actors by developing knowledge production and policy analysis as
a preferred action repertoire. In terms of resource allocation, this led to employing
additional staff who are primarily recruited for their expertise rather than their
records as activists, and with a clear mandate to increase policy expertise and
search for alternative funding sources in order to reduce their dependency on a
major donor, including the state.
Even though it is not a straightforward evolution and while it remains a hotly
debated topic within most organisations, the growing interest of CSOs and
NGOs for policy analysis has long-term practical implications on state–society
relationships within the French context, and more specifically on NGOs and
CSOs themselves. It confirms the continued blurring of frontiers between policy
254
insiders and policy outsiders, as well as the state’s ability to effectively structure
and organise private interests’ participation and contribution to the development
of policy analysis (Dupuy and Halpern, 2009). This in turn led to a renewed
perspective in French policy studies on the role of CSOs and NGOs in policy-
making.
A renewed perspective on the role of CSOs and NGOs in policy-making
The debate about how to examine state–society relationships gained a new

momentum following what is alternatively referred to as a ‘governance turn’ in
policy studies (Gaudin, 2007; Commaille, 2010).16 In the French context this led
to a shift from ‘classic’ approaches to public policy towards a more dynamic, multi-
level and less state-centred understanding of policy-making. In this perspective,
CSOs and NGOs are not considered as policy outsiders, but as non-state actors
that, together with public and private actors, contribute to shaping the policy
process across levels of government and through hybrid action repertoires. By
contrast to the ‘French model of public policy’ in which institutionalisation
dynamics relied primarily upon sectorisation mechanisms, it provides empirical
evidence of bottom-up processes of policy change as well as it demonstrates the
role of hybridisation and recycling mechanisms in shaping long-term policy
developments. In this perspective, less emphasis is put on the role of ideas and
socio-political structures, and more so on policy outsiders, conflicts, innovations
and continued negotiations throughout the policy process (see also Lolive, 1999;
Barthe, 2006; Callon et al, 2013). In addition, increased attention is given to non-
organised forms of policy-making, to the introduction of participatory devices
(Bacqué et al, 2005),17 to the emergence of lay expertise (Boussaguet, 2008;
Fromentin and Wojcik, 2008) and to innovations in governance (Boy et al, 2012).
This changed perspective on the policy-making process highlighted the
growing competition between these organisations and ‘ordinary citizens’ as
policy analysis increasingly relied upon lay expertise. The case of anti-poverty
policy is particularly representative of such attempts to transform policy targets,
users and beneficiaries into policy actors in order to generate alternative forms
of expertise than that produced by socio-economic actors, CSOs and NGOs,
and in some cases, the administration itself. Historically, the need to increase the
socio-political inclusion of the poorest sections of society and their contribution
to policy-making played a critical role in the development of major NGOs such
as ATD Quart-Monde, Fondation Armée du Salut and so on (Carrel and Eme,
2007; Cefaï and Gardella, 2011; see also Paugam and Duvoux, 2008). Recently,
however, formal access to policy-making was given to policy beneficiaries, first as
an experimentation, and following the 2008 law generalising the active solidarity
income and reforming insertion policies, as an institutionalised source of expertise
and knowledge about policies in this field. In this context, the National Council
of Policies against Poverty and Social Exclusion (Conseil national des politiques
de lutte contre la pauvreté et l’exclusion sociale [CNLE])18 formally recognised the
255
expertise of policy beneficiaries in combination with other sources of expertise

(for example, unions, governmental agencies, NGOs, or local authorities) (Weltin,
2013). At first, the creation of the so-called ‘eighth college’ directly challenged
the role played by NGOs and CSOs until then as civil society’s representatives.
But over time, it also led these organisations to strengthen their role as policy
producers by allocating additional resources (for example, human resources, the
production of studies) to the production of policy knowledge in their respective
field and to further distinguish themselves from other participants to the CNLE.
Conclusion
The role played by NGOs and CSOs in the development of policy analysis
evolved considerably over the recent period. This has confirmed the need to go
beyond a simplistic distinction between policy outsiders and policy insiders in
order to fully grasp the numerous ways in which these organisations contribute
to policy-making in the French context. Examples from various policy domains
confirm that they have developed an increased interest in policy analysis as a
strategic action repertoire. This shift is made material by the reallocation of internal
resources, the creation and/or the strengthening of umbrella organisations and the
diversification of influence-seeking strategies that rely upon policy knowledge
and expertise. In other words, these organisations can no longer be categorised as
either service providers or as protesters, but should be seen as directly contributing
to restructuring forms of policy-making as a result of their changed role in the
policy process (Laville and Salmon, 2015).
Yet the role of NGOs and CSOs in policy analysis should not be overestimated.
This cautionary note first derives from their judicial status, which confirms
strong levels of dependency across the voluntary sector upon formal recognition
procedures and public subsidies. As such, only a few large NGOs and CSOs are
able to gain more autonomy, and this is often due to their ability to strategically
combine several action repertoires and to engage into various forms of collective
action at different levels of government in order to multiply sources of funding
and increase their legitimacy. As such the voluntary sector remains characterised
by high levels of inequalities and we suspect that policy knowledge and expertise
plays a critical role in this growing differentiation process. Second, this cautionary
note also derives from the differences observed across policy domains. In this
respect, the development of an actor-centred approach to public policy confirms
– almost by default – the critical role played by classic veto-players (for example,
Grands Corps, professional interest groups) in the French Jacobin policy-making
tradition in relegating NGOs and CSOs to acting as watchdogs, service providers
or in some rare cases as external consultants during policy-making. Only recently,
several cases confirmed the deep mistrust between NGOs and CSOs on the one
hand, and the central and local administrations on the other hand, in their ability
to produce legitimate policy analysis regarding the construction of a major dam
(Sivens) and a new airport (Notre-Dame-des-Landes), in welcoming Syrian
256
refuges or providing migrants wishing to cross the Channel with emergency

accommodation at Calais. In this respect, ‘French exceptionalism’ still has good
days ahead.
Notes
1 By contrast, a lot of work was done on CSOs and NGOs in sociology or in other subdisciplines
within political science, but not necessarily in relationship with the policy process.
2 Religious associations are considered as a major exception and have enjoyed a specific status
since the 1905 Law on secularism.
3 See Chapter Eleven in this volume.
4 See Chapters Four, Seventeen and Eighteen in this volume.
5 Muller suggested a fourth and final circle including other political institutions such as the
Parliament, considered as a weak institution in the context of the French Fifth Republic, and
judicial authorities. See Chapter Eight in this volume.
6 This was also the case of the work done by organisations’ sociologists.
7 Since 1969, FNE has brought together some 3,500 grassroots’ organisations in the field of
environmental protection.
8 It was adopted in 1959 and justified the creation of the Culture Ministry.
9 See also Négrier (2010) on festivals.
10 Other sources of inequalities were highlighted: localisation, political networks, new forms of
cultural expression (for example, comics, photography or street arts) (Arnaud et al, 2015).
11 The support from cultural associations was considered as pivotal in explaining the victory of the
Socialist Party following the 1977 municipal elections and the 1981 legislative and presidential
elections.
12 For a recent work about environmental law, see Doherty and Hayes (2014).
13 A similar evolution was observed in the case of housing following the introduction of the
compulsory right to housing (Weill, 2013).
14 See Chapter Eleven by Hauchecorne and Penissat in this volume.
15 See for example the work about development aid policy (Dauvin and Siméant, 2002).
16 See Chapters Four and Nine in this volume.
17 See Chapter Nine by Blatrix and Gourgues in this volume.
18 CNLE consultative agency was created in 1992 in order to monitor and assess the minimum
income scheme (Revenu minimum d’insertion). It now provides national administrations
with expertise and knowledge about all anti-poverty and anti-social exclusion policies and
programmes.
References
Arnaud, L., Guillon, V., Martin, C. (eds) (2015) Elargir la participation à la vie
culturelle: Expériences françaises et étrangères, Grenoble, Observatoire des politiques
culturelles, www.observatoire-culture.net/fichiers/files/etude_complete_
telecharger_2.pdf.
Bacqué, M.H., Rey, H., Sintomer, Y. (eds) (2005) Gestion de proximité et démocratie
participative, Paris: La Découverte.
Balme, R., Chabanet, D. (2008) European Governance and Democracy, London:
Rowman and Littlefield.
Banaszak, L.A. (2010) The Women’s Movement: Inside and Outside the State, New-
York: Cambridge University Press.
Barthe, Y. (2006) Le pouvoir d’indécision, Paris: Economica.
257
Barthélémy, M. (2000) Associations, un nouvel âge de la participation?, Paris: Presses

de Sciences Po.
Bazin, C., Malet, J. (2015) La France associative en mouvement, Recherches and
Solidarités (13th edn), September, www.recherches-solidarites.org/.
Bereni, L. (2009) ‘Quand la mise à l’agenda ravive les mobilisations féministes:
L’espace de la cause des femmes et la parité politique (1997–2000)’, Revue
Française de Science politique 59(2), 301–23.
Paris: Dalloz.
Boussaguet, L. (2009) ‘Les faiseuses d’agenda: Les militantes féministes et
l’émergence des abus sexuels sur mineurs en Europe’, Revue Française de Science
Politique 59(2), 221–46.
Boussaguet, L. (2013) ‘Agenda/mise à l’agenda’, in C. Achin, L. Bereni (eds)
Dictionnaire genre et science politique, Paris: Presses de Sciences Po.
Boussaguet, L., Surel, Y. (2015) ‘Des politiques publiques “à la française”?’, in
politiques publiques?, pp. 153–84, Paris: Presses de Sciences Po.
Boy, D., Brugidou, M., Halpern, C., Lascoumes, P. (ed) (2012) Grenelle de
l’environnement: acteurs, discours, effets, Paris: Armand Colin.
Callon, M., Lascoumes, P., Barthe, Y. (2013) Acting in an Uncertain World, Boston,
MA: MIT Press.
Carrel, M., Eme, B. (2007) Les communautés Emmaüs dans un monde incertain, Paris:
Union centrale des communautés Emmaüs.
Cefaï, D., Gardella, E. (2011) L’urgence sociale en action: Ethnologie du Samusocial
de Paris, Paris: Bibliothèque du Mauss, La Découverte.
Charvolin, F. (2003) L’invention de l’environnement en France, Paris: La Découverte.
Commaille, J. (2014) ‘Sociologie de l'action publique’, in L. Boussaguet, S.
Jacquot, P. Ravinet (eds) Dictionnaire des politiques publiques, pp. 599–607, Paris:
Presses de Sciences Po, 4th edn.
CPCA (Conférence permanente des coordinations associatives) (2012) La
gouvernance des associations, Paris: CPCA.
Dauvin, P., Siméant, J. (2002) Le travail humanitaire, Paris: Presses de Sciences Po.
Doherty, B., Hayes, G. (2014) ‘Having your day in court’, Comparative Political
Studies 47(1), 3–29.
Douillet, A.C., Faure, A., Halpern, C., Leresche, J.P. (eds) (2012) L’action publique
locale dans tous ses États, Paris: L’Harmattan.
Dubois, V. (2015) Culture as a vocation, London: CRESC, Routledge.
Dupuy, C., Halpern, C. (2009) ‘Les politiques publiques face à leurs protestataires’,
Favre, P. (ed) (1992) Sida et politique: Les premiers affrontements (1981–1987), Paris:
L’Harmattan.
Fillieule, O. (2002) ‘Local environmental politics in France: The case of the
Louron valley, 1984–1996’, French politics 1(3): 305–30.
258
Fromentin, T., Wojcik, S. (eds) (2008) Le profane en politique: Compétences et

engagements du citoyen, pp. 211–36, Paris: L’Harmattan.
Gaubert, M. (2014) La Fédération Addiction, l’expertise renouvelée: Mémoire de master
politiques publiques de santé, Grenoble: Université Pierre Mendès-France, p 139.
Gaudin, J.P. (2007) Gouverner par contrat: L’action publique en question (2nd edn),
Gilbert, C., Henry, E. (eds) (2009) Comment se construisent les problèmes de santé
publique?, Paris: La Découverte.
Halpern, C. (2016) ‘Environmental and energy policy in France’, in A. Appleton,
R. Elgie, E. Grossman, A. Mazur (eds) Oxford Handbook of French Politics, Oxford:
Hayes, G. (2002) Environmental Protest and the State in France, New York: Palgrave
Macmillan.
Hayward, J. (2007) Fragmented France, Oxford: Oxford University Press.
Hayward, J., Watson, M. (eds) (1975) Planning, politics and public policy: The British,
French and Italian experience, London: Cambridge University Press.
Hazareesingh S. (1994) Political traditions in modern France, Oxford: Oxford
University Press.
Hély, M. (2009) Les metamorphoses du monde associative, Paris: Presses universitaires
de France.
Huyard, C. (2009) ‘Who rules rare disease associations?’, Sociology of Health and
Illness 31(7), 979–93.
INSEE (Institut national de la statistique et des études économiques) (2010) Vie
associative, INSEE première 1327, December, Paris: INSEE.
Israël, L. (2009) L’arme du droit, Paris: Presses de Sciences Po.
Jobert, B., Muller, P. (1987) L’Etat en action, Paris: Presses universitaires de France.
Kitschelt, H. (1986) ‘Political opportunity structures and political protest’, British
Journal of Political Studies 16, 57–85.
Kohler-Koch, B. (2008) ‘Civil society in EU governance: a remedy to the
democratic accountability deficit?’, Concepts & Methods 4(1), 3–6.
Lascoumes, P. (1994) L’eco-pouvoir: Environnements et politiques, Paris: La
Découverte.
Lascoumes, P., Le Bourhis, J.P. (1997) L’Environnement ou l’administration des
possibles: La création des Directions Régionales de l’Environnement, Paris: L’Harmattan.
Lascoumes, P., Le Galès, P. (2007) ‘Understanding public policy through its
instruments’, Governance 20(1), 1–21.
Laville, J.L., Salmon, A. (2015) Associations et action publique, Paris: Desclée de
Brouwer.
en France’, in O. Giraud, P. Warin (eds) Politiques publiques et démocratie, Paris:
La Découverte.
Lolive, J. (1999) Contestations du TGV Mediterranée, Paris: L’Harmattan.
259
Lorrain, D. (2005) ‘Urban capitalisms: European models in competition’,

International Journal of Urban and Regional Research 29(2), 231–67.
McBride Stetson, D., Mazur, A. (1995) Comparative State Feminism, London: Sage.
Mouchard, D. (2002) ‘Les mobilisations des “sans” dans la France contemporaine:
L’émergence d’un “radicalisme autolimité”?’, Revue française de science politique
52(4), 425–47.
Muller, P. (1984) Le technocrate et le paysan, Paris: Economie et humanisme, Les
éditions ouvrières.
Muller, P. (2009) Les politiques publiques (8th edn), Paris: Presses universitaires
de France.
Négrier, E. (ed) (2010) Les publics des festivals, Paris: France Festivals and Michel
de Maule.
Nelkin, D., Pollak, M. (1981) The atom besieged, London: Siege.
Ollitrault, S. (2008) Militer pour la planète, Rennes: Presses Universitaires de
Rennes.
Paugam, S., Duvoux, N. (2008) La régulation des pauvres, Paris: Presses universitaires
de France.
Pinson, G. (2009) Gouverner la ville par projet, Paris: Presses de Sciences Po.
Revillard, A. (2009) ‘Le comité du travail féminin et la genèse d’une politique
d’égalité professionnelle en France (1965–1983)’, Revue Française de Science
Politique, 59(2), 279–300.
Rosanvallon, P. (2004) Le modèle politique français, Paris: Le Seuil.
Saurugger, S. (2007) Democratic ‘misfit’? Conceptions of civil society participation
in France and the European Union, Political Studies 55(2), 384–404.
Spannou, C. (1991) Fonctionnaires et militants, Paris: L’Harmattan.
Szarka, J. (2001) The shaping of environmental policy in France, New York: Berghahn
Books.
Tchernonog, V. (2012) ‘Le secteur associatif et son financement’, Informations
sociales 172(4), 11–18.
Tilly, C. (1986) La France conteste de 1600 à nos jours, Paris: Fayard.
Weill, P.E. (2013) ‘Le droit au service des personnes défavorisées?’, Gouvernement
et action publique 2(2), 279–302.
Weltin, L., 2013, ‘De public à acteurs, transformation du rôle des pauvres dans
les politiques sociales’, Mémoire de master Villes, territoires, sociétés, Grenoble:
Université Pierre Mendès-France.
Wilson, F.L. (1983) ‘Les groupes d’intérêt sous la cinquième république’, Revue
Française de Science Politique 33(2), 220–54.
260
FIFTEEN
Trade union expertise in

public policy
Sophie Béroud and Jean-Marie Pernot
The role of French trade unions in public policy has not been discussed much
in the academic literature in the fields of political science or industrial relations.
This lack of interest is due to several factors. First, the French labour movement is
considered unusual in that it is strongly marked by ideologies and very divided. In
international comparisons, it is usually described as atypical, being characterised by
a remarkable ability to mobilise the population despite a low rate of unionisation
(Phelan, 2007; Milner and Mathers, 2013). Second, from 1970 to 2000, authors
who studied the role of trade unions in public policy often used an approach based
on neo-corporatism, applying the theory of macro-level political exchange. These
analyses focus on configurations of interests at the national level, forms of social
arrangements, and dynamics of social pacts. Authors who use this framework
may examine Spain or Italy, for example, but they usually ignore France, where
relations between the ‘social partners’ seem unstable and nationwide social pacts
are rare. Finally, most studies of the French labour movement since the 1990s focus
on its difficulties and examine union organising practices and the mechanisms
through which workers may join or at least support unions.
Studies on countries other than France point to trade unions’ ability to take
on a ‘governmental’ or ‘public administration’ function (Ewing, 2005), which
entails influencing the orientation of public policy and also contributing to the
application of public policy measures. The way unions tend to adhere to the
framework of debate set up by governmental institutions and other stakeholders
– especially associations – leads to thoughts their relative autonomy on their
capacity to produce their own expertise.
Investigating this question can yield new insights into the French situation. The
labour movement in France is characterised by a small proportion of members
– only 7 to 8 per cent of wage earners – and many internal divisions. The
historical origins of these splits lie in ideological differences (between socialists
and communists within the General Confederation of Labour [Confédération
générale du travail (CGT)], and later on between the CGT and the Workers’
Force [Force ouvrière (FO)] [see Box in this chapter on Trade unions in France]),
in religious differences (the Catholic confederation Confédération française des
travailleurs chrétiens [CFTC]), and in differences between categories of workers (the
French Confederation of Management – General Confederation of Executives
[Confédération française de l’encadrement – confédération générale des cadres (CFE-CGC)]
is specific to managers and professional staff).
261
Trade unions in France

Today there are eight national organisations: the General Confederation of Labour (CGT),
the French Democratic Confederation of Labour (Confédération française démocratique du
travail [CFDT]), the General Confederation of Labour – the Workers’ Force (CGT-FO),
the French Confederation of Management – General Confederation of Executives (CFE-
CGC), the Catholic confederation (CFTC), the National Union of Autonomous Unions
(Union nationale des syndicats autonomes [UNSA]), the Solidarity union (Solidaires) and the
Unitary Trade Union Federation (Fédération syndicale unitaire [FSU]). The first five are
officially recognised as representative of employees on the national level in all sectors.
The CGT is the largest confederation. Founded in 1895, it was initially led by
revolutionaries. After the First World War, there was a split between reformists and
revolutionaries; the latter created a separate organisation called the Confédération
générale du travail unitaire (General and unitary confederation of labour). After the
CGT reunified in 1936, the majority of its members were communists; they controlled
the organisation after the Second World War. Most of the reformists left; some founded
FO in 1948, while others became unaffiliated. Teachers set up the Federation of National
Education (Fédération de l’éducation nationale [FEN]) in 1948. The FEN split into
two organisations in 1993: the National Union of Autonomous Unions (UNSA) and
the Unitary Trade Union Federation (FSU), a more militant organisation. The Catholic
confederation, the CFTC, was created in 1919; a reform in ideology resulted in the
abolition of all reference to the Christianity in 1964. The organisation then became the
French Democratic Confederation of Labour (CFDT), although a minority maintained the
traditional CFTC, keeping the old name and an explicitly Catholic ideology. The CFDT
became more radical after 1968 and played a leading role in mass protest movements
throughout the 1970s. The organisation changed its orientation at the beginning of the
1980s and became clearly reformist. Several federations left the CFDT and created a
group of unions called Solidarity, Unity, Democracy (Solidaires, unitaires, démocratiques
[SUD]), which linked up with various unaffiliated unions to form a new organisation
called Solidaires in 1998. Like UNSA, Solidaires has few members and, on the basis of
the results of employee elections, it has not been officially recognised as representative
on the national level. Finally, the CFE-CGC was founded in 1944; its membership is made
up of a small minority of professional staff and managers.
The most important factor who limit the French labour movement’s ability to
influence public policy is linked to the ambiguity and opacity of relations between
political parties and trade unions, especially on the left of the political spectrum.
Due to the tradition of union independence instituted by the Charter of Amiens
in 1906,1 relations such as those that were traditional in social democratic settings
or in the British Labour Party never existed in France, where trade unions do
not overtly influence the programmes of political parties.
This factor contributes to the common notion that trade unions influence
policy only by contributing to mass protests. However, French unions have
developed expertise in certain areas of public policy and they have participated
262
Trade union expertise in public policy
in the design of policy measures. Today, along with employers’ organisations, they
play a role through their participation in the jointly-managed institutions that
run essential elements of the welfare state: health insurance, retirement pensions,
unemployment insurance and job training. This role has brought trade unions to
develop expertise in these areas, even though the state predominates in defining
the orientations of these institutions.
The first part of this chapter examines the history of the French labour
movement from its beginnings to see how it has envisaged its relations with
government and its modes of intervention in government. We show that proposals
elaborated by trade unions have had a strong influence on the design of public
policy at certain key moments in social history. The second part of the chapter
examines the mechanisms devised by the two largest confederations, the CGT
and the CFDT, for gathering information and carrying out analyses; we also
look at how these organisation have become involved in government decision
making in some areas.
An influence linked to certain key historical events

The development of trade union expertise in economics: from the first labour laws
to the movement for central government planning (1890 to 1940)
The CGT’s first years of existence, from 1895 to 1914, were marked by bitter
confrontation with the state. Nonetheless, the CGT’s revolutionary syndicalism
produced new knowledge of the manual workers’ situation. At that time, the
CGT did not have the means to carry out research, which was certainly not a
major preoccupation. Most important were the activities of a few individuals,
such as Alphonse Merrheim, secretary of the metalworkers’ federation, who was
passionately interested in research. The journal La vie ouvrière, founded in 1906,
was revolutionary syndicalism’s organ of reflection. It was designed primarily to
disseminate propaganda rather than to evaluate public policy, but the publication
also reacted to the policy proposals submitted to the parliament and discussed in
the press. For example, during the debates on the creation of retirement pensions
for manual workers that took place between 1901 and 1910 (Renard, 1992), the
CGT argued against building up reserves to fund pensions and against a retirement
age of 65, claiming that it would create a system of ‘retirement for the dead’.
After the First World War, the CGT adopted a modern form of reformism
based on three principles: negotiation of wages through collective agreements,
basic social rights including social security and economic reforms (such as
nationalisation of certain industries). In its 1918 programme, the CGT pushed
for creation of a National Economic Council to study social issues and facilitate
discussion among stakeholders (Lefranc, 1967; Chatriot, 2002). The CGT
produced arguments in favour of a system of social insurance (Assurances sociales) on
the occasion of parliamentary debates preceding the vote on the 1930 law. Over
the following years, a faction in favour of central government planning emerged
263
within the CGT. They felt that the paradigm of a planned economy could be
used to oppose older forms of economic liberalism, which had foundered after
the 1929 economic crisis, and to combat the rise of fascist neo-corporatism.
The group that favoured central planning within the CGT was made up of
trade unionists and also intellectuals who sympathised with the trade unions.
The confederation’s centre for worker education, the Confederation Centre for
Workers’ Education (Centre confédéral d’éducation ouvrière [CCEO]), directed by
Georges Lefranc, was the rallying point for this group (Poggioli, 2008; Lefranc,
1966). Some members of this group followed the lead of René Belin, a prominent
member of the faction, by actively collaborating with the Vichy regime in 1940.
Even though they were few in number, this betrayal by intellectuals close to the
labour movement contributed to FO’s perennial mistrust of intellectuals.
In the 1930s, political considerations dominated public policy debate.
Within the labour movement, politics took on an international dimension: the
revolutionary CGTU was largely subservient to the Comintern; the reformist
wing of the CGT, led by Léon Jouhaux, was active in the International Federation
of Trade Unions (IFTU), which inspired most of its economic programme. The
IFTU collaborated with economists of the Labour and Socialist International
(LSI), experts from the International Labour Organization (ILO), and economists
from the League of Nations. A lecture series given in Zurich in 1930 and 1931
formulated proposals for economic reforms that were taken up by several national
trade unions, including the CGT: these included demands for a 40-hour 5-day
work week, paid holidays and mandatory schooling (FSI, 1931, cited in Pernot,
2001). Within the favourable context of the 1936 Popular Front, these demands
were taken up by the government itself.
Union involvement in consolidation of the welfare state (1945–47)
French unions were much involved in government action just after the Second
World War. The CGT and the CFTC were fully-fledged members of the National
Council of the Resistance (Conseil national de la résistance), and they pushed for
application of the Council’s programme by the three-party coalition government.
At the CGT convention of 1946, Jouhaux presented the confederation’s economic
programme, which included nationalisation of key sectors and central government
planning for the economy. The first draft of the Monnet Plan was considered
‘a solid programme of reconstruction, modernisation and development of the
essential foundations of the French economy’ (Narritsens, 2005). The CGT
played a key role in some sectors such as energy. Coal mines were nationalised
and the national electricity company EDF (Electricité de France) was set up. Both
were headed by Marcel Paul, minister of industry and secretary of the CGT’s
federation of electrical workers (Gaudy, 1996), a double role that reveals a certain
blurring of the boundaries between party and union among Communists. Union
involvement in policy making waned after the war: the Communist Party left
the coalition government in 1947.
264
Throughout the existence of the central government Plan, union experts

were active as members of its Commissions. In 1946, 18 Commissions for
modernisation were set up: the CGT participated in 17 of them and the CFTC
in seven. In 1947–48, the CGT was evicted from the boards of directors of several
public enterprises, but it was not evicted from the Plan (Hayward and Watson,
1975). Pierre Le Brun, a reformist who had originally been an engineer, became
head of the CGT’s economics bureau in 1946. Before the Second World War,
he had participated in the movement for central government planning. The
collaborators he chose for the economics bureau were not Communists for the
most part, but they had decided to stay in the CGT after the FO split off. Little
by little, as members of the Plan Commissions were replaced, the CGT withdrew.
Some Communist CGT members participated in teams involved in developing
the Plan under the leadership of Jean Monnet (Fourquet, 1980).
A new era of union influence after the 1981 election of François Mitterrand
In 1968, as a result of mass social mobilisation, a number of demands that unions

had been putting forward for a long time were finally satisfied. In particular, unions
gained the right to be present in the workplace. A little later, other union-backed
institutional reforms were enacted. These included changes in the system of job
training, new forms of decision-making through systematic collective bargaining
in public enterprises (called contrats de progrès). However, union influence depended
on a favourable balance of power created through strikes and demonstrations. At
the beginning of the Pompidou presidency, from 1969 to 1972, Jacques Chaban-
Delmas and Jacques Delors espoused a ‘new society’ based on social dialogue
and employee participation in company governance; however, their objective
was not to pursue trade union goals but rather to disarm the labour movement
by satisfying some of its demands.
It was only in 1981, however, with the election of François Mitterrand, the first
socialist president of the Fifth Republic, that the labour movement became truly
involved in government and that it was in a position to influence policymakers.
The political configuration changed radically between 1981 and 1984, when
Pierre Mauroy was prime minister. Leaders from the CFDT who were also active
members of the Socialist Party were appointed to positions in some ministries,
including the Ministry of Labour, headed by Jean Auroux. CFDT activists also
actively participated in the creation of the Ministry of the Rights of Women,
working together with Yvette Roudy who was a major figure of the women’s
movement in the 1970’s. Jeanette Laot, for instance, played a pivotal role within the
CFDT in linking the fight for abortion rights with that for wage equality. Similarly,
the four Communist ministers – in charge of the civil service, transportation,
health and job training – chose collaborators from the CGT. Such positions for
union activists in ministries have been so rare under the Fifth Republic that
analysts call attention to them (Mathiot and Sawicki, 1999). Furthermore, the
governments formed by Prime Minister Mauroy sometimes pursued trade union
265
goals and even adopted trade union analyses. Of course, this was the case for the
Auroux laws, which were shaped by the CFDT’s ideas on employee participation
in company decisions (Tracol, 2009). It was also the case for the 1983 reform of
the civil service; in accord with proposals formulated by the CGT federations
of local government and public hospital workers, this reform applied not only
to central government civil servants but also to local governments and public
hospitals. Last but not least, it applied to the Roudy law (1983) which addressed
discriminations against women in accessing work and the gender wage gap. The
trade union’s expertise on this particular issue was pivotal in order to promote a
gender perspective in public policies targeting private companies.
Generally, however, union political influence during this period was weaker than
in the years just after the Second World War. Opportunities for such influence did
not arise again, even at times when the Socialist Party was in power. The labour
movement was profoundly destabilised by the crisis in mines and steel, the rise
in unemployment, and the austerity measures adopted in 1984. At that point,
some members of the CGT, particularly those involved in the confederation’s
economics bureau, tried to design proposals for industry – on the national and
company levels – based on alternative criteria for management (Huiban, 1984;
Lojkine, 1996).
The irrelevance of trade union expertise in a climate of economic liberalism
Like the Mauroy governments, the Jospin government (1997 to 2002) had
ministers who were socialists, communists and ecologists, but it was more distant
from trade unions than at the beginning of Mitterand’s seven-year term of office.
Although the CFDT favoured reducing work time, the confederation showed little
enthusiasm for the Aubry laws instating a 35-hour work week. The CFDT wanted
the reform to be enacted through company-level collective bargaining rather than
through legislation. The CGT did not succeed in stopping deregulation of the
energy sector and it was also unable to prevent the separation between Réseau
Ferré de France, that controls train track, and the train company SNCF, a reform
that was carried out under the aegis of a Communist minister of transportation.
Gender equality provides another interesting example. By contrast to the role
they exerted in the case of the Roudy law (1983), trade unions didn’t influence
the development of the Gender Parity Law (2000) on equal access for men and
women to electoral mandates even though they welcomed it. In the early 2000s,
gender parity was a hotly debated topic within the union, only to promote
women’s access to management positions in public and private organisations
(Silvera, 2006).
Privatisations carried out under the Jospin government widened the gap
between the CGT and the government, and that gap was to grow wider still in
coming years. Along with other unions, the CGT participated in the national
pensions council (Conseil d’orientation des retraites), created in 2000 to monitor
developments in the retirement system, but the CGT has opposed all of the
266
retirement system reforms enacted since 2000. Even when the CGT proposed
alternative measures, they had no impact on public policy, with the single
exception of a system of extra compensation for employees who have worked
under physically harmful conditions, a system that was initially proposed by the
CFDT. In recent decades, the CFDT has supported several policy reforms that
its unionists felt they had inspired or that seemed to be in keeping with CFDT
positions. Concerning changes in the retirement system, the health system, and
government institutions, the CFDT styles itself as a partner of the state and
generally chooses to support reforms in order to influence policy orientations,
at least at the margin. Through its links with intellectuals who are close to the
‘second left’, which favours decentralisation of government, the CFDT has had
access to certain think tanks (for example, the Fondation Saint Simon in the
1990s, Terra Nova and La République des idées since 2000), but the organisation’s
influence on government seems very limited.
A limited production of analyses of public policy

Our summary of the history of union influence on public policy poses the
question of trade unions’ capacity to maintain links with political parties in power,
but above all it poses the question of their capacity to formulate proposals based
on their own analyses. French trade unions have little to do with the university
academics, despite the creation of the Labour Institutes (Instituts du travail) under
the leadership of the university professor Marcel David in the 1950s (Tanguy,
2006). Unions tend to rely on their own rather inadequate internal resources, as
shown by the examples of the two largest confederations, the CGT and the CFDT.
The Economic and Social Research Institute (Institut de recherches économiques et
sociales [IRES]), created in 1982, is an atypical institution, since it is common to
several trade unions and its financial resources are public.
A certain degree of internal independence at the CGT
The development of studies on economic policy was institutionalised in 1946

with the creation of the Confederation Centre for Economic and Social Studies
(Centre confédéral d’études économiques et sociales [CCEES]). This body represents
the CGT in certain organisations like the Plan and also, later on, at the National
Council on Statistics (Conseil national de la statistique [CNS]) created in 1972 for the
national statistical institute (Institut national de la statistique et des études économiques
[INSEE]). Many members of the CGT’s delegation to the Economic and Social
Council (Conseil économique et social [CES]) were from the CCEES. The first
studies carried out by the CCEES concerned the Monnet Plan (1947–53), the
Marshall Plan (1948–52), and the Franco-American Blum–Byrnes agreement
(1946). Subsequently, the CCEES published many studies on trends in wages for
manual workers, purchasing power, consumption and household budgets. The
CCEES undertook a series of international studies and published discussions
267
of macroeconomic policy that were used by the confederation (Magniadas,

2015). The CCEES published extensive analyses of the European Coal and
Steel Community, of the Common Market, and of European and international
economic initiatives. The CCEES is also an important resource for the CGT’s
training centre for unionists, where its members teach courses on Marxist
economics (Ethuin and Yon, 2014).
In 1982, the CGT set up a research institute, the Union Institute for Social and
Economic Research (Institut syndical d’études et de recherches économiques et sociales
[ISERES]), alongside the CCEES, at the same time as the government set up
an institute common to all the major trade unions (see below): the IRES. In
the climate of euphoria that followed the left’s accession to power, the ISERES
adopted an ambitious programme and recruited young researchers. In the 1990s,
the ISERES began to seek European research contracts to finance its activities. A
decline in CGT resources, in addition to management errors, finally brought the
CGT to close the ISERES at the beginning of the 2000s, some 20 years after its
creation. Throughout its existence, one of the institute’s main roles was publication
of an annual report on the economic and social situation, under the supervision
of Henri Jacot. After the disappearance of the ISERES, the CGT did not create a
new research institute; cooperation with researchers, in particular with economists
(Lebaron, 2008) but also with sociologists, results from individual contacts.The
CGT’s economics bureau, the CCEES, has a certain independence within the
confederations central structures as a result of the internal configuration of power.
At first, this independence was largely due to the personalities of Pierre Le Brun,
a member of the CGT governing body. This independence was reaffirmed and
protected by Benoît Frachon, who wanted to show that the CGT was not obliged
to tow the Communist Party line. At certain moments of tension, the CCEES
was called back into line. Nonetheless, its independence was genuine, and it was
maintained by Le Brun’s successors: Jean-Louis Moynot, member of the CGT
governing body from 1967 to 1981, and Gérard Alezard from 1982 to 1995.
Moynot was among the CGT leaders who wanted to open up the organisation
to non-dogmatic communist ideas. He was in favour of putting forward ‘counter-
proposals’ designed to deal with problems in companies or industrial sectors, rather
than simply opposing government measures. In 1979, he resisted the seizing of
power at the CGT by hardline Communists. Tensions were manifest during the
1979 labour conflict over steel production in Lorraine when Moynot and his
colleagues criticised the tactics of the CGT’s metallurgy federation. Disagreements
became more pronounced on the question of Europe. Moynot, who disapproved
of the Soviet intervention in Afghanistan and the CGT’s position on Solidarnosc
in Poland, stepped down, along with three other members of the CGT’s governing
body. Alezard, who succeeded Moynot, was much more ‘orthodox’, but, toward
the end of his mandate, he joined other top leaders in pushing for faster reform
of the CGT. At the end of the 1990s, Jean-Christophe Le Duigou, head of the
economics bureau, was labelled a ‘moderniser’ by opponents within the CGT.
268
Knowledge for training union activists at the CFTC and CFDT
The history of the CFTC – and of its successor, the CFDT – is quite different
from that of the CGT but it is similar in some ways. Before the Second World
War, members and leaders paid little attention to public policy: the programme
adopted by the CFTC in 1935, called a ‘Plan’ at the time, was very different from
the agenda of the CGT and had nothing to do with central government planning.
In keeping with ideas on the reconstruction of the social order contained in the
Quadragesimo anno papal encyclical of 1931, the CFTC favoured the principle of
subsidiarity of higher levels of decision making to local levels and the principle
of a minimal role for the state, limited to guaranteeing a balance between social
forces. Nonetheless, the CFTC stated that the study of economic questions was
imperative for trade unions. During its first convention in 1920, the CFTC
condemned materialist doctrines and pleaded for changes in work conditions. The
organisation did not push for radical change in the social order but rather sought
to harmonise the social order with the Christian humanist values. Consequently,
the CFTC attached an intrinsic value to research in and of itself, in particular to
analyses of the economic context of union action.
After the Second World War, a CFTC fringe group, led by Paul Vignaux,
spearheaded debate within the Catholic organisation. It influenced a ‘minority’
within the CFTC, made up of industry or regional federation leaders who wanted
the confederation to abandon the links with the catholic Church. Economic
questions were central to the Cahiers (‘Notebooks’) published by this faction
of the CFTC, while the confederation itself intervened mainly in reaction to
government initiatives. Little by little, under the influence of dissidents, the
CFTC came to approve the concept of central planning; as early as 1950, General
Secretary Marcel Gonin insisted on the importance of investment as a means of
giving a social and democratic orientation to the economy. In 1955, pressure
from the dissident ‘minority’ increased. In 1959, the CFTC convention adopted
a report on democratic government planning that constituted a turning point
in the Catholic confederation’s concept of the economy. The report contained
a significant number of proposals designed to increase government control
on investment. In 1961, in response to proposals for economic cooperation
formulated by the reformist faction of employers’ organisations, the CFTC
affirmed that its aspirations were incompatible with maintenance of the capitalist
structures of companies. At the 1961 CFTC convention, a report on ‘wage policy’
stated that collective bargaining alone could not regulate wages sufficiently and
that a national governmental income policy was needed. Economic analyses of
planning and social democracy thus played an important role in 1964 when the
majority of the CFTC membership transformed the organisation into the CFDT.
Throughout the history of the CFTC-CFDT, it attached great importance
to training for militants, particularly as a means to understanding the economy
(Henry, 1956). Because of the need to constantly update information and analyses,
the confederation created an institute for union research and training (Institut
269
confédéral d’étude et de formation syndicales). About 20 regional institutes (called

Écoles normales ouvrières) were set up to organise training sessions. The content of
training courses changed with the transformation of the CFTC into the CFDT,
but training continued to be based on a two-fold goal: self-realisation through
general education and improving the effectiveness of union action. It was felt
that activists should be capable of understanding both company strategies and
the orientations of public policy. After 1964, the focus switched to democratic
central planning. Albert Détraz, who was in charge of the confederation’s political
bureau, exhorted members to become ‘technically revolutionary’. The expression
implied that education and training were imperative for an organisation whose
members had to be effective activists indeed in order to compete with the CGT,
which had many more members.
During the 1960s and until 1973, René Bonéty headed the CFTC-CFDT’s
economics bureau, one of six confederation bureaux. He called on the expertise
of another department of the confederation called the Bureau of Research and
Economic Action (Bureau de recherches et d’action économique [BRAEC]). The
BREAC was headed by Jacques Delors until 1962, when he was put in charge
of social affairs at the General Commissariat of the Plan. At that time, the
BRAEC published many documents that included economic analyses but were
not restricted to economics, since the BRAEC’s field of competence was broader
than that of the CCEES. Generally, the CFDT focused less on economics than the
CGT. Because of its particular relationship to Marxism, the CFDT systematically
carried out political analyses: a political bureau was created in 1961; social analysis
became its forte and May 68 enhanced this role. Between 1973 and 1982, the
economics bureau was headed by Michel Rolant along with experts such as Jean-
Pierre Oppenheim, Jean-Pierre Aubert and Hubert Prévost. These men were
more influenced by Marxism than other members of the confederation and their
positions were contested, especially after the outbreak of the economic crisis in
1973. Economic analysis was mobilised in internal struggles over the CFDT’s
orientation for the years to come. The strategy adopted at the end of the 1970s
focused on action in the workplace rather than political activities; it was based
on an interpretation of the crisis as being, not cyclical, but structural, or even a
crisis of civilisation.
Pierre Héritier replaced Rolant in 1982 as head of the economics bureau.
Like his predecessor, his was an explicitly non-dogmatic approach to Marxism.
Héritier named a team of economists, but he also chose sociologists. Research and
statements on public policy at this period concerned subjects that were national
in scope, such as hygiene, security and conditions in the workplace. During the
2000s, the CFDT launched several action research programmes, notably on the
transformation of work.
270
Weak links with universities, offset by a common research institute
In all the areas where unions intervene regularly (job training, wage equality,
institutions that are jointly run by employers and unions, and social insurance
in general), confederations try to develop analytical frameworks to guide them
in their role as managers. Unlike unions in others countries, they do not have
permanent links with university research centres. The only area in which there
is long-term cooperation with academic researchers is gender issues. Sociological
surveys on the role of women in trade unions are regularly led, in connection
with the CFDT, the CGT, the FSU and Solidaires.
The unions can also rely on the IRES, created in 1982 ‘to serve trade unions
that are representative of employees’. The IRES was created at the same time as
the OFCE, which is designed to inform public debate on government policy,
and Rexecode, a research institute intended to serve employers’ organisations.
The creation of these institutes was inspired by the Lenoir Report, written
in the framework of the eighth Plan. The report recommended fostering the
production of economic and social information by institutions other than the
national statistical office, the INSEE, which practically had a monopoly at the
time.The IRES has a two-fold structure. On the one hand, it has a main office
which initially was supposed to be mainly a documentation centre but which
developed research activities, especially when Jacques Freyssinet was the director.
On the other hand, the IRES finances research carried out by the different trade
unions. The unions choose the subjects in which they are interested and submit
their reports to the IRES, which makes them available to all the unions. Initially,
the goal was to encourage unions to develop their own research capacities in order
to improve ‘social partnership’ (although the term was not in use in the IRES’s
early years), at a time when economic restructuration threatened to increase
tensions in labour relations. Up to 2006 the IRES grew in notoriety and in size,
in terms of both the number of researchers and the number of studies carried
out by the six trade union organisations that were members of the institute.2
The IRES is the only institution that groups together all of the major French
trade union confederations. This cohabitation has persisted despite many conflicts
over the last 30 years. The confederations receive some financial support for
their activities from participating in the IRES, but the sums in question are
small compared to the other subsidies they receive in various forms from the
government. The main benefit for participating unions is that financing from
the IRES has guaranteed a certain continuity in the research carried out by
confederations; within the smallest organisations, such as the CFTC or the
CFE-CGC, research would probably have disappeared altogether due to budget
restrictions stemming from their small memberships. In addition, the IRES
has heightened confederations’ awareness of the importance of international
comparisons.
271
Links between the production of information and the internal

dynamics of trade unions
Trade union confederations are not the only organisations to produce information
and expertise on public policy. Certain federations that have succeeded over
decades in organising a large proportion – even a majority in some cases – of
workers in their sectors of activity can shape the regulation of industrial relations
and the orientation of public policy. This is the case for the CGT’s mining and
energy federation in the field of nuclear energy. It is also the case for teachers’
federations: there was a single organisation, the FEN, in the past; it has now split
into UNSA-éducation and the FSU. Since the Second World War, teachers’ unions
have participated in designing education policy, through their participation in
consultative bodies (such as the Superior Council of Teaching (Conseil supérieur de
l’enseignement) and through the positions they take on reform proposals (Robert,
2004). For a long time, these unions were inspired by the 1947 Langevin-
Wallon Plan, which advocated a school system designed to achieve a more
egalitarian society, in keeping with the programme of the National Council of
the Resistance. Despite some differences between categories of teachers (primary
school, secondary school), the unions that made up the FEN elaborated their
own plan for making education more democratic. They managed to influence
the Haby reform in the 1970s who instituted a single type of secondary school.
However, it was the teachers’ federation of the CFDT, the General Trade Union
of National Education (Syndicat général de l’éducation nationale [SGEN]), which
grouped all categories in a single organisation, that inspired another major
innovation in education, the priority education area (Zone d’éducation prioritaire
[ZEP]), a mechanism for enhancing the resources of schools in deprived areas,
instituted by the Mauroy government in 1981 (Robert, 2004, 122–8). In this
case, union expertise was used to analyse a problem (inequalities in education)
and to elaborate concrete measures that directly influenced government action.
This episode destabilised the SNES, the FEN’s union of secondary school
teachers. The capacity of teachers’ unions to draw up programmes for coherent
alternative reforms of the educational system, concerning pedagogical objectives
and the content of school programmes, has waned considerably over the last 20
years. One reason for this is the divisions between categories within the main
federation, the FSU: each national union – primary school, secondary school,
university – defends its own membership. Thus, teachers’ unions have tended
to limit themselves to defensive reaction to reforms proposed by government,
and they have few resources available to come up with viable counter proposals.
This example from a specific sector helps to reveal the kind of expertise that
unions need. Schmitter and Streeck’s (1999) distinction between the logic of
influence and the logic of membership is helpful in discerning what a union
needs to faithfully represent its membership and what it needs to participate in
shaping public policy. These two types of action are interconnected, as the example
of teachers’ unions shows. Manufacturing federations also need to produce
272
guidelines for policy in their specific sectors, in order to orient their company level
organisations, who are too taken up with daily struggles against redundancies and
delocalisations to have a broad picture. Such a goal has inspired the way the CGT
has approached the question of work in recent years, trying to apply the practical
know-how that union activists have acquired as elected or appointed members
of the Hygiene, Safety, and Work Conditions Committee (Comité d’hygiène, de
sécurité et de conditions de travail [CHSCT]). This know-how can be fashioned into
a body of knowledge that can be used to shape political orientations and to train
unionists. On the basis of the idea that the issue of the link between employee
health and work conditions can be used to reform union practices and promote
unionisation, the CGT has formed a ‘work and emancipation’ taskforce, made
up of researchers, practitioners and union activists, which has published several
documents. Inspired by experiences such as the struggles of workers exposed
to asbestos, the group has developed positions which include demands for
compensation for victims and demands for adding more work-induced illnesses to
those officially recognised by public institutions. The leaders of the group feel that
it has influenced some parts of the third national Plan for Workplace Health (Plan
Santé au travail), in force from 2015 to 2019. However, there have been internal
disagreements over the idea of making this subject a priority for the confederation
and devoting a large share of the organisation’s intellectual and training resources
to it. The capacity to produce independent policy orientations in this area, based
on solid arguments and prepared in cooperation with professionals (especially
doctors in charge of workplace health) without becoming subordinated to those
professionals, is not considered a priority by all the CGT’s leaders. Some prefer
to stick to more traditional demands.
These tensions reveal unions’ difficulties in proving to members that participating
in consultative bodies – as opposed to decision making bodies – in certain areas
of public policy is worthwhile. Such participation may result in cooptation of
union activists, but it is also a source of information and knowledge.
Conclusion
Our study of periods when the French labour movement has been able to
influence public policy leads us to several conclusions. The first is that unions have
constantly tried to produce analyses of their social and economic environment,
even when such analyses were carried out by only a small group of union
leaders or experts. This preoccupation goes beyond the laws and reforms that
directly confront unions, whose research agendas are not dictated by the public
policy agenda. Although French unions are sometimes depicted as weak and
highly dependent on government (Howell, 1998), their capacity to maintain
independence vis-à-vis the political sphere has been linked to a certain capacity
to produce independent expertise. A second conclusion is that union influence on
public policy has been strongest at exceptional times, marked by major political
events (the Second World War) or sweeping social protest movements (1936, May
273
68). Their ability to put their ideas forward successfully thus seems to be linked to
a favourable balance of power established through struggles rather than through
institutions designed for consultation or negotiation (notably institutions that
are jointly managed by unions and employers’ organisations). The third and last
idea concerns the role of production of knowledge. While one of its goals is to
orient public policy in a direction that the unions want, it also has internal uses.
By developing expertise in particular areas, unions not only show government
that they are legitimate political actors, they also reinforce members’ perception
of their legitimacy as spokespersons and representatives.
Notes
1 Adopted by the 1906 convention of the CGT, the Charter of Amiens proclaimed the total
independence of the trade union from the socialist party and defined a two-fold role for the
union: to act to change workers’ present situation; to prepare for worker’s social, economic,
and political emancipation.
2 Representatives of the five trade union organisations that are officially considered to be
representative of employees on a national level sat on the IRES’s board of directors, along with
representatives of the federation of public school teachers, FEN. Today, the composition of the
board of directors is the same, except that the FEN has been replaced by UNSA.
References
Chatriot, A. (2002) La démocratie sociale à la française: L’expérience du Conseil national
économique (1924–1940), Paris: La Découverte.
Ethuin, N., Yon, K. (eds) (2014) La fabrique du sens syndical, la formation des
représentants des salariés en France (1945–2010), Broissieux: Le Croquant.
Ewing, E.K. (2005) ‘The function of Trade Unions’, Industrial Law Journal 34(1),
1–22.
Fourquet, F. (1980) Les comptes de la puissance, Paris: Encres, Éditions Recherches.
FSI (Fédération syndicale internationale) (1931) Aux prises avec le chômage et la
crise économique, Amsterdam: FSI.
Gaudy, R. (1996) Et la lumière fut nationalisée, Paris: VO Éditions.
Hayward, J., Watson, M. (eds) (1975) Planning, Politics and Public Policy: The
British, French and Italian Experiences, Cambridge: Cambridge University Press.
Henry, F. (1956) ‘La CFTC et l’éducation ouvrière’, in C. Blondel, J. Bussienne,
L. Carliner (eds) Culture ouvrière et action syndicale, Paris: Éditions du Cerf.
Howell, C. (1998) ‘Virtual unionism in France’, in H. Chapman, M. Kesselman,
M. Schain (eds) A Century of Organized Labor in France, New York: St Martin’s
Press.
Huiban, J.-P. (1984) ‘La contre-proposition industrielle comme élément de
stratégie syndicale (1973–1980)’, in M. Kesselman, G. Groux (eds) 1968–1982:
Le mouvement ouvrier français, crise économique et changement politique, pp. 295–313,
Paris: Les éditions ouvrières.
Lebaron, F. (2008) ‘A contre-courant: Le renouveau de l’économie critique en
France depuis 1995’, in B. Geay, L. Willemez (eds) Pour une gauche de gauche,
pp. 177–94, Broissieux: Le Croquant.
274
Lefranc, G. (1966) ‘Le courant planiste dans le mouvement ouvrier français de

1933 à 1936’, Le mouvement social 54, 69–89.
Lefranc, G. (1967) Le mouvement syndical sous la Troisième république, Paris: Payot.
Lojkine, J. (1996) Le tabou de la gestion, la culture syndicale entre contestation et
proposition, Paris: Atelier.
Magniadas, J. (2015) ‘La CGT et l’économie’, Analyse et documents économiques
120, 5–9.
Mathiot P., Sawicki S. (1999) ‘Les membres des cabinets ministériels socialistes
en France (1981-1983): recrutement et reconversion’, Revue Française de Science
politique 49(1), 3–30.
Milner, S., Mathers, A. (2013) ‘Membership, influence and voice: A discussion
of trade union renewal in the French context’, Industrial Relations Journal 44(2),
122–38.
Mioche, P. (1987) Le Plan Monnet: Genèse et élaboration (1941–1947), Paris:
Publications de la Sorbonne.
Narritsens, A. (2005) ‘Programme économique ou programme d’action, le
tournant du 30ème congrès de la CGT’, in E. Bressol, M. Dreyfus, J. Hedde,
M. Pigenet (eds) La CGT dans les années cinquante, pp. 403–13, Rennes: Presses
Pernot, J-M. (2001) Dedans dehors, la dimension internationale dans le syndicalisme
français, PhD thesis, Université Paris X – Nanterre.
Phelan, C. (ed) (2007) Trade Union Revitalisation: Trends and Prospects in 34
Countries, Oxford: Peter Lang.
Poggioli, M. (2008) ‘Le planisme à la CGT: Les origines d’une refonte syndicale
au tournant du Front populaire’, Cahiers d’Histoire103, 27–40.
Renard, D. (1992) ‘Une vieillesse républicaine? L’État et la protection sociale
de la vieillesse, de l’assistance aux assurances sociales (1880–1914)’, Sociétés
contemporaines 2(10), 9–22.
Robert, A. (2004) ‘Jeux croisés des syndicats d’enseignants face aux réformes
et projets de réforme (1944–2000)’, in J. Girault (ed) Les enseignants dans la
société française au XXe siècle, itinéraires, enjeux, engagements, pp. 111–32, Paris:
Publications de la Sorbonne.
Schmitter, P., Streeck, W. (1999) ‘The organization of business interests: Studying
the associative action of business in advanced industrial societies’, Max-Planck-
Institut für Gesellschaftsforschung Discussion Paper 99(1), 1–95, Cologne: Max-
Planck-Institut für Gesellschaftsforschung.
Silvera, R. (2006) ‘Le défi de l’égalité hommes/femmes dans le syndicalisme’,
Mouvements 43(1), 23–9.
Tanguy, L. (2006) Les instituts du travail: La formation syndicale à l’Université de 1955
à nos jours, Rennes: Presses universitaires de Rennes.
Tracol, M. (2009) Changer le travail pour changer la vie: Genèse des lois Auroux,
Paris: L’Harmattan.
275
Part Five
Academic policy analysis
277
SIXTEEN
Economics and policy analysis:

‘from state to market’?
Daniel Benamouzig and Frédéric Lebaron
Introduction
Deeply rooted in a French political tradition, economic expertise is expressed in
both widespread and diverse ways. One general feature that has remained persistent
is undoubtedly its strong yet complex and shifting relationship with the state:
economic expertise has remained largely public. As in other western countries,
market mechanisms at both the macroeconomic and microeconomic levels now
play a greater role in public policy. Economic experts, however, still primarily
work within or, increasingly, with national and transnational public bureaucracies.
Although there is a greater reference to markets, economic expertise has remained
strongly connected to bureaucratic action.
In this chapter, we analyse three different periods which characterise different
historical configurations. We examine the role of French economists in the state at
different levels; although their role was rather national at the outset, it has become
more international and transnational in recent times. In the post-war decades,
economic expertise was mainly developed within the state and promoted either
a micro or macroeconomic state-centred approach. French engineer-economists
were trained in Grandes Écoles and employed as civil servants in the administration
where they acted as experts. They played an important role in the monetary and
fiscal policy at the national level, that is, in macroeconomics. They were also key
players at the sectoral level and participated in the early development of health
economics, in close interaction with national planning (Benamouzig, 2005). The
case of health is a particularly good test of the sectoral variations in a broader
process because it was long perceived as resistant to economic reasoning. From
the 1970s, the academicisation and internationalisation of economics developed
following a clear shift toward microeconomic reasoning. Moreover, the influence
of pro-market expertise in the public sphere – at both the national and sectoral
level – also increased, as evidenced by the aborted project of privatisation of
the French healthcare system. Finally, the reinforced internationalisation of
economic expertise in recent times has paved the way for pro-market policies.
The international spread of economic ideas in public decision-making has
also developed alongside the rise of new technical bureaucracies devoted to
economic expertise. At the transnational level, this new economic bureaucracy
has expanded and now covers both national policies and international bodies.
279
The Europeanisation of the fiscal and monetary policy illustrates this trend which
can also be observed at the sectoral level. The rise of national agencies working
transnationally in health technology assessment illustrates this process: while new
bureaucracies seem to differ from the former national ones, they nonetheless share
some common traits. This, in turn, has called into question the role economic
expertise may play in modern transnational societies.
We will therefore focus on whether this evolution is specific to French history
or, rather, is part of a wider international trend, embedded in global forms of
economic expertise. Given the importance of transnational bureaucracies in recent
times, a progressive convergence towards a rather general model can be expected.
However, path-dependency processes may explain other national features as well.
A state-centred economic expertise

In this section, we focus on the importance of the national state in the field of
economic expertise in France. Specifically, we analyse the first primarily national
orientation until the 1970s and highlight the role of engineer-economists.
The fiscal and monetary policy is a key element of a ‘Keynesian’ economic
construction; we therefore focus on this policy to show the centrality of the
national state in the development of economic expertise in France.
State economic policy: money, budget and performance
Economics emerged in the seventeenth century as a pure ‘cameral science’1 within

the royal court. Although it initially revolved around powerful political actors, it
soon became an activity practiced by politicians, high-ranking civil servants and
various industrial actors during most of the nineteenth century (Perrot, 1992).
From the end of the nineteenth century, a tradition of engineer-economists
emerged. They were trained within scientific and technical schools (Grandes
Écoles) and participated in major innovations across various fields (Etner, 1987).
As in many other countries, economic knowledge in France developed within
the state. It provided a foundation and acted as a point of reference for new types
of public policies such as ‘macroeconomic policies’ that emerged in the 1940s.
In parallel, political economy and, subsequently, economics were initiated and
established as scientific disciplines. They enjoyed a close connection with public
policy and a certain degree of scientific autonomy. ‘Academically’ speaking,
economic expertise was first developed in France in the nineteenth century in the
so-called Grandes Écoles (such as L’école Polytechnique) rather than in universities.
Generally speaking, scientific innovation did not occur in universities until the
second half of the twentieth century, as evidenced by the lack of possible career
options for Léon Walras in France (Dumez, 1987).2 As such, unlike in other
countries, the interplay between policy and economic knowledge can hardly be
reduced to a simple interaction between academics and economic policy in France,
280
Economics and policy analysis
because it was first and foremost an institutional process which transformed both
the state and the forms of knowledge.
The interplay between mathematics, especially probability and statistics,
engineering sciences, and social issues best illustrated by the work of Maurice
Allais3 in the post-war decades, led to the emergence of a ‘modern-style’
mathematical and statistical economics integrated within French Administration
(Desrosières, 1993). From the end of the war, economists were present within
the state; they produced and analysed national accounts and economic models
and forecasts by adapting them to specific sectoral conditions. This was behind
the success of the discipline in the policy sphere. Economists were primarily
located at the École nationale de la statistique et des études économiques, and then at the
École Polytechnique. Both centres were closely connected to a recently established
technical department at the Ministry of Finance, that is, the Direction de la Prévision
(as described by Fourquet, 1980). Between the 1950s and the 1960s, economics
emerged and expanded under the impulse of actors such as Edmond Malinvaud
who was the director of the National institute of Statistics and Economic Studies
(Institut national de la statistique et des études économiques [INSEE]), the School for
Advanced Studies in the Social Sciences (École des Hautes Etudes en Sciences Sociales)
and Professor at the Collège de France (Lebaron, 2000).
Monetary and fiscal doctrines drove economic reasoning within the state.
During the twentieth century, this became a global trend in the day-to-day
functioning of the modern state. As in other countries, the economic reasoning
mobilised in France was strongly based on the current state of the political
economy. Drastic changes were observed over the years when macroeconomic
models emerged; the first were changes in intellectual technology (Armatte,
2010) in the 1930s, and then in how the models were applied in the 1940s.
Largely based on statistics, this macroeconomic expertise was also restructured
and applied to different sectors of public policy such as industry and trade. It was
also applied to social policy, especially education and health, but was ultimately
challenged by more ‘microeconomic’ and managerial types of reasoning in the
1970s. Since the Second World War economic expertise has become a universal
tool for public policy.
Over the centuries, monetary doctrines have traditionally developed at the
intersection of treasuries and central banks. During the twentieth century and
after the Second World War in particular, the interaction between these doctrines
and the academic world became increasingly important. Most of the features
of the ‘Banque de France doctrine’ appear to have remained relatively stable
(especially with regard to its anti-inflationist stance and its pragmatic conception
of monetary order). After the Second World War, however, the institutionalisation
of macroeconomic thinking and action led to the transformation of its conditions
of application. There was a strong dependence on the central bank, defined at
the time as an administration within the state. Monetary policy was inspired by
the then dominant views in macroeconomics, that is, a sort of pragmatic and soft
Keynesianism implemented by financial civil servants (inspecteurs des finances). This
281
vision of macroeconomics emphasised the use of econometric models produced

and managed by engineer-economists (at the INSEE and the Direction de la
Prévision [DP]) which provided the basis for ‘fine-tuning’ the economic policy.
During the same period, a parallel process affected budget policy. During
the reconstruction years, the state was the central actor and planning (with the
Plan Commission [Commissariat Général du Plan]) was one of its major tools. In
particular, the state set long-term production and modernisation objectives. It
sought to modernise the economy and make it more productive, in terms of
both quantity and quality.
After the Second World War, many academic actors (either scientists working
in the Grandes Écoles, the National Centre for Scientific Research (Le Centre
national de la recherche scientifique [CNRS]), or university professors) already
had relationships with policy actors and institutions at various levels. This is
evidenced for instance by François Perroux, a catholic intellectual who was then
a prestigious Professor of Economics (at Sciences Po, then at the Collège de France).
He regularly voiced his strong opinions on policy issues, including during the
Vichy regime years, and, years later, at the French Economic and Social Council
(Conseil économique et social [CES]). Among his notable contributions was his
involvement in the creation of national accounts after the Second World War.
He also participated in the creation of a critical theory of development which
served as a foundation for various expertise activities undertaken by his followers
in newly independent countries in the 1960s and 1970s. This will be further
discussed in the sections below.
Academic economists nevertheless played a minor operational role: they were
involved in the production of a general discourse about the economy which
was diffused by politicians, especially in the 1960s. This discourse then began
to be challenged by much more market-oriented and pro-business views. From
the outset, the French conception of planning was both ‘administrative’ and
‘market-oriented’. The borders between more interventionist and more liberal
(free-market) conceptions within the French administration were somewhat
blurred. As planning declined, Europeanisation and globalisation became more
constraining. Academic economists were thus regularly called upon to propose
guidelines to enhance the position of France and Europe on the global market.
Health economics and planning at national level
At the sectoral level, the emergence of health economics in France illustrates the
crucial role of administrative and public bodies in the post-war period. Health
economics had long been inhibited in France by several factors. While some of
these factors were common across developed countries, others were more specific
to France. First, the establishment of medicine as a profession enabled physicians
to protect their activities from external economic regulation by both the state
and the market. Second, the social security system created in the aftermath of
the Second World War allowed a consensual increase in public spending devoted
282
to healthcare, and this did not require stringent economic regulation. Third, a
pervasive ideology of modernisation and progress in the healthcare sector paved
the way for the strong support of expensive innovations which did not require
economic assessment. However, recurrent financial tensions were increasingly
taken into account. Health economics expertise was initially promoted by
the state administration rather than by academic experts (Benamouzig, 2005).
The first attempt to develop health economics in France resulted from the
connection between the national planning system and a few physicians working
in the administration. Working alongside the Plan Commission established
in 1946 to implement the national Plan, a small group of experts comprising
physicians and statisticians were charged with the mission of developing health
consumption national accounts in 1954. This was the first attempt to develop
‘medical economics’ within a wider macroeconomic framework. Interestingly, no
economist was part of this team, which helped quantify the ‘need’ for hospitals
and health spending for decades. The data gathered by statisticians and interpreted
by physicians such as Henri Péquignot and Georges Rösch were consistently
aggregated. This was in line with the consensus in the medical profession in
the post-war period: while medical consumption was accounted for globally, it
was perceived as legitimate for public spending to increase alongside physicians’
revenues. As such, professional practices remained unquestioned.
In the late 1960s, the introduction of cost–benefit analysis in health economics
met with some resistance. In January 1968, the Ministry of Finance launched a
national operation known as Rationalisation of Budgetary Choices (Rationalisation
des Choix Budgétaires [RCB]) which sought to adapt the American Planning
Programming Budgeting System to the French administration. Engineer-
economists initially trained at École Polytechnique had long used cost–benefit
analyses in some sections of the French administration such as transport and
defence planning. In the early 1970s, the health sector was targeted as an
experimental field in which to deploy these methods. A special unit led by high-
ranking officials and closely connected to the Secretary of State for Health was
set up. Over the next few years, many programmes were assessed; these included
perinatal care, cervical screening, psychiatric care and vaccination. While the
use of cost–benefit analysis was viewed as technically innovative, it proved to be
politically dubious. The methods used tended to oversimplify possible options.
In several cases, the very criteria chosen for the assessment were inappropriate.
The quantification of the so-called ‘price of life’ was even viewed as scandalous. A
new generation of economists working in the administration eventually rejected
cost–benefit analyses. Younger experts trained primarily in the USA were highly
reluctant to use the pre-existing economic instruments for planning such as
macroeconomic models and cost–benefit analyses. They found the methods
inappropriate at a time when fluctuation and crisis were common. A pro-market
international wind was blowing from the United States and it reached the French
circles of economic expertise.
283
While there was a slight increase in the role of academics during this period, it
remained limited compared to the role of administrative experts. Some academics
acted as advisors and others were involved in the training of public experts.
However, this occurred in Grandes Écoles rather than in the universities. The shift
to more clearly market-oriented policies would later be accompanied, and even
determined, by the greater input from from academic experts.
The emancipation of academic economics and the rise of market-

oriented policies
In this section we describe the evolution of public policies toward more market-
oriented policies. We also highlight the growing importance of academic
economists as a global profession driven by market-oriented policies in a complex
causal relation (Fourcade, 2009). In the field of health policies, this trend can be
observed both in economic policies and at the sector level. It does not imply a
shift of economic expertise from the public to the private sector; rather, it implies
a diversification of expertise with a growing role of external academic specialists
within policy decision processes.
A pro-market economic policy?
From the 1970s onwards, a more pro-market economic policy was progressively
taking root in France. It was, first and foremost, the result of changes among
political and administrative elite groups who saw this orientation as the best
response to globalisation. This evolution is illustrated by the historical emergence
of political actors such as Raymond Barre, Jacques Delors and Michel Rocard
who based their careers on ‘economic pragmatism’ and especially on ‘budgetary
rigour’ (‘rigueur budgétaire’) and/or ‘austerity’ (‘austerité’). Some of these actors
(for instance Raymond Barre who was a professor at Panthéon-Sorbonne and
author of a book that was famous in the 1960s–70s) explicitly referred to academic
knowledge which had been developing in the US around monetarism and ‘supply-
side economics’. However, they abandoned neither their political and economic
pragmatism nor their Keynesian references. These actors perceived public debt as
dangerous and urged a return to the ‘fiscal balance’ doctrine. In parallel, neoliberal
doctrines were largely diffused in the political field and in the media after 1984.
At the end of the 1970s, monetarism became much more influential among
high-ranking civil servants as well as in some academic circles. More dramatic
changes occurred in the monetary and fiscal policy, especially after 1983–84.
The changes were driven by the triumph of ‘competitive disinflation’ and the
pro-European orientation of the socialist government and its liberal successor
in 1986. After 1988, ‘alternances’ after the general elections did not change this
general orientation that was deeply rooted in a new alliance between politicians,
senior civil servants and financial and industrial actors. From the 1990s, this shift
was associated with the greater role played by academic economists in certain
284
sectors of economic policy. This role, however, has remained modest compared
to the central role played by Grands Corps and traditional state actors.
The domination of the Bundesbank in Europe and the major changes observed
within financial and monetary administrations across the world (the International
Monetary Fund, the World Bank and the Organisation for Economic Co-
operation and Development [OECD]) help explain the rapid shift toward the
orthodox doctrine of ‘competitive disinflation’ in the 1980s. This move was
initially promoted by high-ranking civil servants such as Jean-Claude Trichet.
The changes were supported by a network of academics such as the influential
Christian de Boissieu (Panthéon-Sorbonne) who worked in connection with the
Treasury and the central bank.
With the growing role of public independent structures such as the Court of
Audit (Cour des comptes) in the assessment of budget policies, the coalition between
different groups of civil servants (financial inspectors [inspecteurs des finances],
civil administrators [administrateurs civils] and statisticians-economists [statisticiens-
economistes]) has remained central within the French administration. However,
the role of macroeconomic forecasts and statistics seems to have declined in the
long run: competing predictions of GDP growth continue to be important in
the political debate, especially during fiscal negotiations between ministers and
the head of government, or in electoral debates. However, discussions on the
macroeconomic effects of fiscal stimulus, and more generally, of fiscal decisions
have been bypassed by the general emphasis on fiscal stability since the second
half of the 1980s.
The rising role of academic economists
During the 1960s, French economic departments within universities became

more important and gained in legal autonomy. Their evolution favoured ‘modern’
research activities. In particular, this evolution favoured scientific production, that
is, the publication of research articles with some mathematics and statistics in
peer-reviewed journals. This brought it closer to the Grandes Écoles tradition of
engineer-economists which had occurred in the 1980s and 1990s and had been
driven by various academic and administrative actors such as Henry Guitton and
the French Assocation of Economics (Association française de science économique) as
well as by a new generation of academics in major universities such as the Sorbonne
University (Paris-1) and the University of Nanterre. During this period, a clear
shift toward mainstream economics occurred within universities and the CNRS
where critical and heterodox conceptions had been of great importance in the
1970s. A political economy close to the social sciences has remained present in
these institutions to this day.4 In February 1981, Raymond Barre, a professor of
economics who had also been Prime Minister since 1976, decided to create new
departments in order to provide a more diversified macroeconomic expertise for
the state as well as for non-state actors: thus were established the French economic
observatory, the Observatoire français des conjonctures économiques (directed by Jean-
285
Marcel Jeanneney who had been close to General de Gaulle), the Institute for
Economic and Social Research (Institut de recherches économiques et sociales) (related
to workers’ union confederations) and Rexeco (which later became Rexecode and
was related to business organisations, see Delmas, 2007). This paved the way for
a certain diversification and a greater role for think tanks. It may also have led to
the relative decline of macroeconomic knowledge in the hierarchy of legitimate
cognitive tools and a preference for microeconomic models which were also
developed within the state.
In the following years, new economic styles gained importance within the
university setting. New classical economics influenced macroeconometric
forecasting and heterodox theories were far less present in public policy circles.
The shift was internal in part; socialist and critical academic economists rapidly
shifted to more liberal conceptions of economic policy after 1984, as evidenced
by the example of Dominique Strauss-Kahn. This is even more clearly illustrated
by Strauss-Kahn’s former colleague Denis Kessler who might be seen as a symbol
of this evolution following his shift from a critique of inequality to an active
promotion of pension funds. As Denord (2007) has described, the rise to power
of the right wing in 1986 greatly revived the French liberal tradition. Maurice
Allais’ Nobel prize in 1988 was interpreted as the consecration of this part of
French academic production which was closely associated with the evolution of
large public companies (Finez, 2015). Economists such as Christian de Boissieu
played a role in this general shift. The shift involved a rapid decline of left-wing
and Keynesian viewpoints rather than the genuine emergence of a new type of
economist; this economist had actually already been long present.
Health economics offers another good example and exemplifies the general
trends introduced in this section.
International health economics, from state to market?
Health economics began to refer to the market in the mid-1980s in the quest
to professionalise and globalise health economics (Benamouzig, 2009). The first
international congress of health economics in Tokyo in 1973 did not target
academic economists alone. Most participants were specialised civil servants
working in national public administrations. The French delegation in particular
comprised experts working either in the Plan Commission or in the Ministry
of Finance. Although John Kenneth Arrow’s very famous and seminal paper on
asymmetries of information was being discussed at the international level, it was
barely taken into account in France (Arrow, 1963). Only a couple of French
economists even mentioned it in the 1970s. The situation changed dramatically
in 1979 when a second international congress of health economics took place in
Leyden, the Netherlands. Many more academic economists participated in the
congress and met foreign colleagues on professional grounds. North American
economists increasingly used microeconomics to analyse patients’, physicians’ and
insurers’ behaviour. For instance, the Canadian economist Robert Evans presented
286
a remarkable paper that grasped the main features of any national health system
in a simple analytic framework. He was able to enhance the strategic behaviour
of ‘agents’ within the framework. This model made international comparisons
easier. It was used as an analytical tool across countries. It also elicited new
conceptions of health reforms: by comparing various national systems according
to a common set of principles, one could imagine possible transitions from one
system to another. Moreover, it became possible to observe different national
logics that had previously been perceived as insular and idiosyncratic. Evans used
economics to make structural reforms feasible and comparable.
Other American economists were more genuinely committed to pro-market
reforms. Working behind the scenes (when Ronald Reagan was President),
Professor Alain Enthoven promoted the American model of health maintenance
organisations at the international level. This model was supported by the
Republicans from the 1970s onwards. These organisations, which were often for-
profit companies, provided managed care for health insurance. Travelling across
Europe after the Leyden congress, Enthoven met influential national experts
and decision-makers. He met with some success in several countries such as the
UK, because of the Nuffield Trust, for instance. In France, he got on well with
an academic economist trained in the USA, Robert Launois, who was fond of
microeconomics and had an interest in pro-market political reforms. In 1985,
a think tank recently launched by a few French private companies, La Boétie
Institute, promoted pro-market ideas in view of the upcoming elections. They
asked Launois to write a book on market competition in healthcare (Giraud and
Launois, 1985). Advertised by the Institute, the project received much attention.
It marked a turning point in the history of the French healthcare system. Against
all odds, a few physicians showed interest in this free-market option. On the
one hand, the anti-state perspective appealed to right-wing physicians who had
long been hostile to public regulation. On the other, left-wing physicians were
interested in a stronger integration of medicine in collective practice. Naturally,
traditional frontiers were blurred. While the reform was perceived as globally
unconvincing, its microeconomic foundations were unanimously acclaimed.
Along with other innovative academic research, microeconomics became a
lingua franca in health economics. Although market competition proved relatively
incompatible with the national healthcare system, reformers viewed the spread
of microeconomics as crucial.
During this second period, academic economists played a greater, albeit limited,
role in policy-making in France. It remains, however, difficult to quantify this
role. This change was largely driven by a shift toward microeconomics applied,
in particular, to sectors such as health which had long been perceived as relatively
inadequate sectors for economic reasoning. In macroeconomics, the shift was
internal and consisted in a rapid decline of Keynesian and Marxist references. It
led to a rapid change in favour of more liberal views.
The strongest influence, however, was associated with the transnationalisation
of the economic policy which accelerated after 1984. This transnationalisation
287
blurred the shift toward a market-oriented policy because it was diffused in a

complex set of bureaucratic institutions.
The transnational bureaucratisation of economic expertise?

In this section, we highlight the internationalisation of economic expertise in a
more recent period (from the mid-1990s onwards). This period is marked by a
growing reference to bureaucratic transnational institutions which tend to promote
or regulate pro-market orientations.
Experts from both the administration and the universities are now present.
They often compete in this new globalised and bureaucratised field of economic
expertise where private actors as such remain marginal. Here again, economic
policy and health economics provide good examples of this evolution.
The international bureaucratisation of economic policy
The Council of Economic Analysis (Conseil d’analyse économique [CAE]) was

established by Lionel Jospin in 1997 and was linked to the office of the Prime
Minister. Although it symbolised a new trend, this trend had actually begun in
the early 1990s. It involved placing greater emphasis on academic expertise in
economic policy-making, from the macroeconomic level to more specific sectoral
issues such as retirement reform. While numerous and undoubtedly influential,
the reports produced by the CAE were still competing with classical administrative
reports from a wide variety of agencies and institutions. Although some of the
economic experts in the CAE were from the financial sector, others were from
the administration (including statistical administration).
This can best be illustrated through two rapid case-studies of the monetary
policy and ‘structural reforms’. European monetary unification provoked a
strong institutional shift which led to the emergence of a more direct presence
of academics in decision-making spheres: in the first ‘council of monetary policy’
of the newly independent Banque de France in 1994, Denize Flouzat incarnated
the growing presence of French academics within decision-making bodies. With
the creation of the European Central Bank (ECB), traditional financial senior
civil servants maintained their leading position.
In 2012, however, Benoît Coeuré, a member of the Polytechnique-ENSAE
and author of several academic books about European economic policy was
appointed to the Executive Board of the Governing Council of the ECB. This
illustrates the growing interest in more scientific profiles in the management of
money at the European level. Although Coeuré received the support of a section
of academia on the occasion of the appointment of a new governor of the Banque
de France in 2015, the head of the state preferred François Villeroy-de-Galhau, a
classical Finance Inspector.
Fiscal stakes have been largely recast as a ‘structural’ issue related to efficiency
and ‘structural reforms’. Following a trend partly initiated by academics, fiscal
288
policy has become less involved with the short-term state of the economy or
with issues pertaining to social justice. Academic actors have helped shape this
new conception of fiscal policy. Actors such as Jean Tirole, Pierre Cahuc, André
Zylberberg and Philippe Aghion still participate in the recurrent debate about
pensions and their structural reforms and labour market reforms. Since the 2000s,
these academic economists have argued in favour of the flexibilisation of the
labour market in line with the recommendations of the European Commission
and the OECD.
This was the case of the ‘Attali report’ of 2007 (Attali, 2008); this brought
together high-ranking civil servants, modernist business leaders and a few reformist
academics such as Philippe Aghion. The report proposed a ‘rupture’ in economic
policies through the acceleration of structural reforms and liberalisation. As the
global financial crisis of 2008 affected some of their propositions, academic
economists expressed their disappointment at the lack of interest shown in their
ideas by political and bureaucratic actors.
In 2008, Nicolas Sarkozy and Christine Lagarde launched a debate on the
measurement of economic performance and social progress. The Stiglitz–
Sen–Fitoussi report produced by members of academia from major American
institutions, including four Nobel laureates, was a major contribution to the
debate (Stiglitz et al, 2009). Although actors from international organisations
and statistical institutions were also present, they had a secondary role which was
demonstrably something new in the French traditional division of labour. The
report proposed major changes, some aspects of which have been implemented
by national institutes of statistics. The report clearly reinforced the position of
academic economists in policy debates in France and strengthened core economic
issues such as: What is wealth? What is performance? What is progress? (Lebaron,
2010).
The references to econometric models and business-cycle debates remain
common in the economic discourse around today’s policies. A typically Keynesian
type of reasoning continues to shape the predictions and analyses on the impact
of a budget on growth and unemployment. Nevertheless, their political effects
seem less significant compared to the emphasis placed on the importance of
taxes expected to limit the country’s competitiveness in global markets. Basic
macroeconomic discourse is now produced just as much by international agencies
such as the OECD and the European Commission as by national actors such as
INSEE or the Treasury. Since 2010, national debates have tended to follow the
critical stance of a section of leading US academic economists as regards European
policy. This has led to the emergence of a huge divide pitting the experts of the
ECB, the European Commission and the Ministries of Finance against a fraction
of leading world economists such as Joseph Stiglitz, Paul Krugman and Thomas
Piketty. This divide shows that macroeconomic reasoning in the international
scientific community is still important. It also reveals the tension between
economic policy criteria promoted by European institutions on the one hand,
and the more global and less consensual academic debates on the other.
289
Health economic bureaucracies at transnational level?
In the 1990s, several public policies reinforced the use of health economics.
However, no structural reform implemented at the time was as comprehensive as
that imagined by Robert Launois a decade earlier. The use of health economics
was rather pervasive in several public institutions. In the public health sector,
the implementation of national ‘plans’ targeting specific diseases and disorders
made it possible to assess the efficiency of such plans using cost–benefit methods.
These methods were also proposed to assess health technologies, especially
in relation to drugs. The pharmaceutical industry invested large amounts of
money to develop ‘pharmaco-economics’ and adapted the good old cost–benefit
analysis to randomised clinical trials. In academia, a national ‘College of health
economists’ was created to adequately address new industrial needs. The College
brought together academic economists, civil servants, professionals working in the
pharmaceutical industry and physicians interested in the field. Some economists
were hired by both the administration and the industry. Their methods were,
however, contentious. Not only were the studies sponsored by the industry rare,
they were also often biased. National civil servants and physicians involved in
the regulation of drugs expressed growing doubts about the soundness of these
methods.
At the international level, the destiny of economic methods did not encounter
a similar stalemate. In 1999, French health economists gazed admiringly at the
National Institute of Clinical Excellence (NICE) established in the United
Kingdom. The Institute had been placed in charge of health technology assessment
in the NHS. According to French economists, this innovation paved the way for
the use of economic methods in decision-making. No institutional equivalent
existed in France then, nor was there any political will to create one. Transnational
dynamics, however, played a decisive role. At the European level, independent
agencies were set up in the 1990s to guarantee the independence and soundness of
decision-making, especially for drug market authorisation and food safety. Other
agencies were also set up to regulate and/or promote specific health policies.
In France, one agency was set up to promote the quality of care. Gradually, its
missions extended to health issues at large. In the 2000s, a new Higher Authority
for Health became competent in health economics. An academic economist,
Lise Rochaix, who had previously worked in the French administration and
in the United Kingdom, was appointed to the board of the authority. A small
department of economics was set up to deliver expertise in ‘medico-economics’.
Parliament acknowledged this competence in 2012 which was then extended
to drug-pricing procedures. This evolution brought the French authority closer
to its British equivalent. Such proximity facilitated horizontal relationships
with a number of similar agencies in Europe such as the German Institute for
Quality and Efficiency in Health Care (Institut für Qualität und Wirtschaftlichkeit
im Gesundheitswesen [IQWiG]) and the Belgian Health Care Knowledge Centre
(Federaal KenniscentrumView on Map [KCE]) among others.
290
Over the years, the work undertaken by these kinds of agencies has gone beyond
the national level. It has involved the sharing of processes across similar bodies
at the European level. The economic guidelines defined by national agencies
and consequently used by the industry for regulatory purposes have been widely
discussed among national institutions and experts. Their respective conclusions
have circulated across agencies. The periodic revisions of the guidelines have been
undertaken alongside benchmarking processes. Incorporated in the day-to-day
activities of national agencies, health economics has become standard protocol,
facilitating routine uses. This new use of much older economic methods has given
rise to ambiguous results. On the one hand, agencies show familiar traits of the
so-called ‘regulatory state’, that is, they are limited to steering functions. On the
other hand, they have gradually acquired some traits of classical bureaucracies such
as independence and formal ways of doing. They have also acquired some level
of hierarchy, between European and national levels for example. This practical
use of health economics has not expanded to the national level; for instance, it
has not affected the historical bureaucracies which have been in existence since
the 1960s and now operate at transnational level along with bureaucratic forms
of organisation. Although health remains a national competence of member-
states in Europe, economic regulation facilitates horizontal relationships between
member states’ agencies. This process differs from classical European Community
integration because national agencies are not associated with an integrated
European body. Rather, they relate with each other at the transnational level.
Health economics has thus been involved in the gradual formation of a new
bureaucratic web connecting independent bodies beyond frontiers.
Conclusion
Over the years, both the content and scope of economic expertise have changed.
The interactions between economics and public policy have become more
complex and diversified. Moreover, the field of economics has become much
more diverse although the dominant academic discourse is highly compatible
with ‘market solutions’. From the 1970s, economics gained autonomy within
universities and higher education institutions. In parallel, the internationalisation
and ‘academicisation’ of the discipline occurred over the same period. This
influenced public policies at both the national and sectoral levels. It occurred
alongside the increased use of academic works in public decision-making. At the
national level, the changes observed in fiscal, monetary or economic performance
policies illustrate the wider transformations that have occurred in the relationship
between economic experts and public policies. At the sectoral level, the case of
the health sector, which has increasingly been influenced by economic references
in the past decades, shows how these general trends have been able to permeate
more specific areas.
Given these dynamics, the competition between senior civil servants, state
economists and academic economists has never really ceased but has become
291
increasingly blurred with the rise of international (especially European) levels

of governance. This governance is characterised by new forms of bureaucratic
integration; while it at times advocates the expansion of market mechanisms, at
other times it does not. These relationships have made the well-known trajectory
‘from state to market’ more complex. The trajectory has characterised the
‘classical’ narrative on the relationship between state, economics and the market
from the 1980s onwards. Judging from recent transformations at the transnational
and international levels, economic expertise has not simply shifted ‘from state
to market’; it has also shifted from the market to new technical bureaucracies
operating at both national and transnational levels. In some cases, this has restricted
market mechanisms. These recent trends associate state economists and academic
economists more than ever; while these economists work together, they also
compete at the very heart of new bureaucratic and transnational arenas. These
trends have simultaneously been observed at different levels and may ultimately
favour the renewed expansion of market mechanisms or the creation of new
bureaucratic regulations that limit market expansion.
Notes
1 A cameral science is a discipline deeply shaped by its connection with a policy sector: see
Chapter Two by Payre and Pollet in this book.
2 This was the case until relatively recently when both institutional traditions began to converge
as illustrated by today’s Centres of Academic Excellence (Toulouse School of Economics and
Paris School of Economics). In parallel, economics developed in the Grandes Écoles, the French
institute of statistics and its extensions, the École des Hautes Etudes en Sciences Sociales, and at the
National Centre for Scientific Research (Le Centre national de la recherche scientifique [CNRS]).
Academic expertise underwent great changes, as did its relationship with policy analysis.
3 The first French man to be awarded the Nobel Prize in 1988 – if we don’t take into account
the French American Gérard Debreu.
4 In the 1950s and 1960s, this tradition was frequently referred to as ‘école française’ or
‘unconventional economics’; it is today defined as ‘heterodox’ economics (Pouch, 2001).
References
Armatte, M. (2010) La science économique comme ingénierie: Quantification et
modélisation, Paris: Presses des Mines.
Arrow, K.J. (1963) ‘Uncertainty and the welfare economics of medical care’,
American Economic Review 53(5), 941–73.
Attali, J. (ed) (2008) Rapport de la commission pour la libération de la croissance française,
Paris: La documentation française.
Benamouzig, D. (2005) ‘L‘État au chevet de l‘économie médicale’, la contribution
du CREDOC’, in P. Bezes, M. Chauvière, J. Chevallier, N. de Montricher, F.
Ocqueteau (eds) L‘État à l'épreuve des sciences sociales. La fonction recherche dans les
administrations sous la Ve République, pp. 88–110, Paris: La Découverte.
Benamouzig, D. (2009) ‘Professionnalisation académique et engagement partisans
des économistes de la santé (1970–1990)’, Sociétés Contemporaines 1(73), pp.
73–95.
292
Delmas, C. (2007) La Chambre de commerce et d’industrie de Paris: Un ordre négocié,

Droit et société 67, 597–613.
Denord, F. (2007) Néolibéralisme version française: Histoire d’une idéologie politique,
Paris: Demopolis.
Desrosières, A. (1993) La politiques des grands nombres, Paris: La découverte.
Dumez, H. (1987) Le cas Walras: L’économiste, la science et le pouvoir, Paris: Presses
Universitaires de France.
Etner, F. (1987) Histoire du calcul économique en France, Paris: Economica.
Finez, J. (2015) Pratiques économiques et pensées du changement dans un service public
marchand. Une sociologie des chemins de fer français aux XIXe et XXe siècles, Thèse
pour le doctorat de sociologie, Université de Lille, 1 novembre.
Fourcade, M. (2009) Economists and Society: Discipline and Profession in the United
States, Britain and France, 1890–1990, Princeton, NJ: Princeton University Press.
et du Plan, Paris: Encres, Éditions Recherches.
Giraud, P., Launois, R. (1985) Les réseaux de soins: Médecine de demain, Paris:
Economica.
Lebaron, F. (2000), La croyance économique: Les économistes entre science et politique,
Paris: Seuil.
Lebaron, F. (2010), La crise de la croyance économique, Broissieux (Bellecombe-en-
Bauges): Le Croquant.
Perrot, J.C. (1992) Une Histoire Intellectuelle de l’Economie Politique, Paris: Éditions
de l’EHESS (L’ École des hautes etudes en sciences sociales).
Pouch, T. (2001), Les économistes français et le marxisme (1950–2000), Rennes:
Presses universitaires de Rennes.
Stiglitz, J., Sen, A., Fitoussi, J.P. (2009) Richesse des nations et bien-être des individus,
Préface de Nicolas Sarkozy, Paris: Odile Jacob.
293
SEVENTEEN
The academic world of French policy studies:

training, teaching and researching
Patrick Hassenteufel and Patrick Le Galès
The gap between the policy analysis developed by policy actors and the existing
academic knowledge on public policies has been identified in previous chapters
as being central in France. This chapter aims at exploring and providing some
explanation of this division. As with the rest of this book it relies upon the
distinction between, on the one hand, policy analysis, defined as ‘applied social
and scientific research – but also involving more implicit forms of practical
knowledge – pursued by government officials and non-governmental organisations
which usually focus on designing, implementing and evaluating existing policies,
programmes and other courses of action adopted or contemplated by states’, and,
on the other hand, policy (process) studies ‘conducted mainly by academics and
relate[d] to “meta-policy” or the overall nature of the activities of the state. It is
generally concerned with understanding the development, logic and implications
of overall state policy processes and the models used by investigators to analyse
those processes’ (Dobuzinskis et al, 2007, 1).
In France this gap is particularly vivid, despite an importation of policy analysis
in the 1960s and 1970s (see Chapter Three) and is illustrated by the absence of
autonomisation of policy analysis in specific academic departments or institutions
(corresponding to public policy schools). French policy analysis has not triggered
the creation of a new disciplinary field. By contrast, policy studies have been
integrated and have contributed to existing academic disciplines, in particular,
sociology, political science and economics. This situation is sometimes obfuscated
by the use of the French expression ‘analyse des politiques publiques’ which refers
more to policy (process) studies in the abovementioned sense than to policy
analysis.
Our hypothesis is that this strong segmentation between expert knowledge of
public policies (policy analysis) and scientific knowledge of public policies (policy
studies) is historically rooted in the power organisation of the French academic
field at large, which has three main characteristics (Heilbron, 2015, 6). The first
one is the centralisation of academic institutions and production in Paris. This
centralisation produces specific centre–periphery relations. However, innovations
are often initiated in the periphery to challenge the centre. The second one is the
strong hierarchy between very selective elite higher education schools (Grandes
Écoles) and universities (open to all students having a high school degree, without
selection). It has several consequences: the reinforcement of centralisation in Paris,
the control of access to academic and administrative (so as political) positions
295
by these Grandes Écoles and a strong competition between them. The third
characteristic is the institutionalised separation between research and teaching.
Much of the research is historically concentrated in specific research institutions,
for social sciences especially the National Centre for Scientific Research (Le Centre
national de la recherche scientifique [CNRS]) and the School for Advanced Studies
in the Social Sciences (École des Hautes Études en Sciences Sociales [EHESS]).
The first part of this chapter develops a long-term historical analysis of the
French higher education system in order to highlight its legacy on the development
of policy analysis and policy studies. This development is analysed in the second
part with two main arguments. First, in relation to the centralisation of elite
Grandes Écoles in Paris, the training of senior civil servants brought together policy
analysis and law. By contrast policy process studies developed at the periphery,
within less prestigious higher education organisations, often in relation to the
CNRS. Second, the academisation process of policy studies in the existing
disciplines (especially political science) has widened the gap with policy analysis.
In our final section we point towards some recent evolutions that may contribute
to partly bridge it.
A long-term separation between the training of elites, universities

and research
The gap between policy analysis and policy (process) studies was explained in
previous chapters by the legacy of administrative science (Chapter Two) and
the domination of state expertise (Chapter Six). In these chapters, with minor
exceptions, university professors are more or less absent from the story, except to
a limited extent in the faculties of law. This section further explores the same line
by focusing on the organisation of the higher education system, with a specific
interest for the main disciplines involved in the study of public policy: sociology,
political science and law (the previous chapter already dealt with economics).
The historical dichotomy between universities and Grandes Écoles
Universities are only one section of the French academic world, and not the
most prestigious one, with limited contribution to the training of state elites.
By contrast, the best students were historically trained in isolation both from
classic academic disciplines and from research within specific French institutions:
the Grandes Écoles. Unsurprisingly, at a later stage, when those students become
leading policymakers, they ignore what kind of academic knowledge is being
produced. They are ‘the experts’, they are the ‘state nobility’ to take Bourdieu’s
famous word (Bourdieu, 1989), they take decisions, why would they need policy
studies? Tellingly, among state elites, a PhD is seen more as a handicap than an
added value. However, significant changes are now taking place as will be shown
in the last section of the chapter. Until recently, no academic institution looked
like a public policy school.
296
The academic world of French policy studies
France has an original higher education and research system (Charle and
Verger, 2007). Historically, in parallel to other European countries, medieval
universities developed, in Paris (the famous Sorbonne), Montpellier, Strasbourg,
Toulouse or Bordeaux among others. Then came the time of the creation of
professional schools, in most cases outside universities. They were particularly
numerous in France, illustrating the inertia of universities and the lack of trust
of the centralising state elites for rebellious and conservative universities. Before
the French Revolution three schools for the training of state engineers were
created: the École Royale des Ponts (then École Nationale des Ponts et Chaussée) in
1743, the École de genie de Mézieres in 1749 and the École des Mines in 1783. This
model, opposed to universities, based on the acquisition of professional skills, a
direct control of the state and the selection of students, was reinforced during the
French Revolution with the creation, in 1794, of the École Polytechnique (training
future state engineers) and the École Normale Supérieure (for the best scientific
and literary students). These elite institutions were labelled Grandes Écoles, in
order to train the elites who would run the country, and build separately from
the universities (the only exception are the Écoles Normales Supérieures structured
around the academic disciplines and training future university professors, in the
same way as high school professors and researchers). The core role of these elite
schools and the suspicion towards universities is symbolised by their abolition
during the French revolution (in 1793). Re-created by Napoleon they were
then strongly controlled by the state, fearing political contestations (which was
the case in 1830 and 1968 for instance). The political control of universities is a
long-term issue. They are under state scrutiny and face constant reform attempts.
This appears in stark contrast with the never-ending development of the
Grandes Écoles (outside universities) during the nineteenth century in relation to
industrial development and economic growth. Notable here were the creation of
the École spéciale de Commerce et d’Industrie (later the École Supérieure de Commerce
de Paris[ESCP]) in 1819 (the first French business school), of the École centrale in
1829 (training engineers for the industry), of the Haute École de Commerce in 1881,
of the École Supérieure d’Electricité (1894) and the École Supérieure d’Aéronautique
(1909). Other business schools were then developed after the First World War.
The whole French education system has progressively become geared towards
selecting the brightest pupils (or rather the more prepared and adapted to the
education system with a high level of social reproduction) for this elitist group of
Grandes Écoles, through highly competitive national exams. With some exceptions
(medicine), the top 5 per cent of the students after baccalauréat (the end of high
school examination) enter competitive intensive special higher education classes
within high schools (classes préparatoires) instead of going to university, in order
to pass the national exams, after two years of preparation, and then enter those
elitist Grandes Écoles.1 Therefore, progressively, they recruited the children of the
Parisian bourgeoisie and a minority of bright young students from all over the
country. By contrast, universities are not allowed to make a selection of their
students, hence a massive gap in terms of prestige and resources.
297
For quite a long time therefore, the university system has not attracted the
best French students. Educated within these elitist small Grandes Écoles, French
scientific, administrative and economic elites have been trained to run the state and
French firms, with the self-confidence of being the chosen meritocratic superior
elites. Interestingly, for many of those schools (with the exception of the Écoles
normales supérieures), the training does not include many contacts with research in
the different academic disciplines. Teaching in these institutions is characterised
by the refusal of specialisation in academic disciplines, the promotion of practical
exercises and evaluation and the valorisation of the acquisition of a general
knowledge distinct from academic research and based on pluri-disciplinarity.
Training in these schools lasts between three or four years and delivers a specific
diploma, not a university degree (until recently). Students of these schools
therefore rarely then go on work towards a PhD, their school diploma being
sufficient to be recruited into top administrative or economic positions. A last
characteristic has to be stressed: the teaching and direction functions in these
schools are held by alumni with careers in the administration or/and in the private
sector, but not in universities. Academic recognition does not play a role and the
main part of the teaching staff of these schools is non-permanent.
These characteristics are particularly relevant for the Grandes Écoles providing
training for public policies and administration. In this perspective, it is important
to mention the creation of the École libre des sciences politiques in 1871 by (minority)
protestant elites. Even if it had a private status it shared the practical orientation
of the elitist public Grandes Écoles, the separation from university faculties and
training based on a flexible combination of different disciplines (mainly law,
history and economics, and ‘modern social sciences’ such as political science and
sociology). The research objectives of the founder of the school, Emile Boutmy,
were very soon abandoned in favour of the preparation for the competitive exams
giving access to upper-level positions in the administration (Favre, 1981). It was
nationalised after the Second World War and transformed into the National
Foundation of Political Sciences (Fondation Nationale des Sciences Politiques)
managing the Institute of Political Studies (Institut d’Études Politiques [IEP]) in
Paris, the successor of the École libre, then known as ‘Sciences Po’, and the new
IEPs created in Bordeaux and Grenoble. The majority of students who passed the
competitive national competitive exam to enter Sciences Po would prepare then,
during their three years curriculum, the even more competitive exam to enter
the National School of Administration (École Nationale d’Administration [ENA]),
also created in 1945 to train the future upper-level civil-servants (see Chapter
Six). Sciences Po, like the ENA, was run by top civil servants on the one hand
(from the Council of State mostly, later also from the Inspection des Finances) and
a small number of university professors in history, political science and sociology.
Sciences Po graduates have for long more or less monopolised access to ENA.
The second main source of state elites involved in public policies are the
scientific Grande Écoles, and most importantly the École Polytechnique, leading to
specialised highly prestigious engineering schools (for instance the École des Mines
298
or the Écoles des Ponts et Chaussées). Graduates from those Grandes Écoles (also from
the National School of Statistics and Economic Administration [École Nationale
de la Statistique et de l’Administration Économique, ENSAE]) in statistics mentioned
in previous chapters) have for decades monopolised the production of expertise
and top jobs in various policy sectors in ministries, agencies and large firms alike.
Within Sciences Po and the ENA law was still present but mostly taught by the
members of the Council of the State, not by university professors. Social sciences
or research about the administration or public policy, not to mention comparison,
were more or less non-existent. French top civil servants were trained to be the
elites, had worked hard in competitive exams with a strong sense of meritocracy,
they run large firms and ministries, they produced expertise, took decisions with
next to no contacts with universities or research centres. They had no clue why
policy studies might be relevant.
The distinct growth of social science research
The dichotomy between universities (with a rather low status, except in medicine,
law and history) and Grandes Écoles is not sufficient to explain the French higher
education system. A third kind of organisation, the CNRS, was set up by the
Front Populaire (left wing government) in the late 1930s and officially created in
1939. One explanation for this creation of a distinct national organisation for
advanced research was the conservatism of universities. Some leading scientists
did not find the right conditions to develop advanced research. After the war,
beyond physics and more classic sciences, the CNRS opened its doors for young
researchers in disciplines that were either underdeveloped or not very popular
within universities, for instance sociology (Mendras, 1995). CNRS would
welcome those young people who were later supposed to get a more prestigious
university job. After 1945, sociology had become marginal in the university system
(except in Bordeaux and Paris). CNRS progressively welcomed a generation of
young people interested in sociology and the change of their society, who would
learn their trade in the US, often benefiting from Ford Foundation scholarships
to go to Chicago, Harvard or Columbia. With or without the US journey, Alain
Touraine, Michel Crozier, Edgar Morin, Henri Mendras to name a few, made
their research within CNRS. Research centres were created such as the Centre
for the Sociology of Organisations (Centre de Sociologie des Organisations) in 1964,
with Michel Crozier and a group of young researchers (Renaud Sainseaulieu,
Pierre and Catherine Grémion, Jean-Claude Thoenig, Jean-Pierre Worms, later
Erhard Friedberg). During the 1960s, until the mid-1970s research funding in
social sciences grew fast and was oriented towards the CNRS and the EHESS
(Heilbron, 2015, 162–6). French social science research developed in CNRS
research centres first in sociology then in political science with a creation of a new
section including the discipline. It is important to note that CNRS researchers
did not have to teach and were isolated from students. Sociology developed in
universities only in the 1970s.
299
By contrast, political science emerged both in Sciences Po and in the faculties

of law. Political science was institutionalised in part in France at Sciences Po,
with the creation of the French Association of Political Science (Association
Française de Science Politique) in 1949, of the French journal of political science,
Revue Française de Science Politique, in 1951, both financed by the FNSP, and the
two first research centres in political science in the mid-1960s, the Centre for
the Study of French Political Life (Centre d’Études de la Vie Politique Française
[CEVIPOF]) and the Centre for International Relations Studies (Centre d’Études
des Relations Internationales [CERI]). Both were to become CNRS centres later. In
accordance to what was mentioned before, researchers within these centres were
not involved in teaching, and then became in charge of basic teaching. Sciences
Po’s teaching was done by this relatively small group of senior professors and a
large number of professionals, dominated by ENA graduates who had become
top civil servants. One would have thought that the close interactions between
public policy practitioners and experts on the one hand, political science senior
professors and a growing body of research fellows would have led to the rise
of policy studies on the one hand, and of policy studies courses to train future
civil servants on the other. That was not the case. First, state administrative elites
providing most of the teaching for the students fiercely resisted any attempt to
include solid social science in the curriculum for decades (including economics).
Several attempts were made by sociologists (Crozier and Friedberg) in the 1980s
to train students with policy studies or organisations’ sociology. That remained
either non-existent or marginal. In the late 1980s, the political science professor
Yves Mény, with CNRS research professor Jean-Claude Thoenig also mobilised
several scholars to provide a strong public policy programme including some of
his British colleagues (Jack Hayward and Vincent Wright). He tried to develop
that both in the PhD programme but also for the students preparing for ENA.
The project was unambiguously rejected. Later, both Erhard Friedberg, then
director of the Centre of Sociologies of Organisations (Centre de Sociologies des
Organisations [CSO]) (a joint CNRS/Sciences Po research centre by then), and
Jean Leca, senior professor of political science, made several similar attempts. All
were rejected. The Council of State and the French elite civil servants would
not accept social science in their training. Even in the 2000s, the rather radical
director of Sciences Po, Richard Descoings, several times started to work on a
public policy school project that would have included policy studies in political
science, sociology and economics. Although a member of the Council of State
himself, he faced systematic opposition and only designed the premise of the
project: state administrative elites still trained future state administrative elites.
Until the 1980s, top students were trained in elitist Grandes Écoles with little
contact with university professors and little access to research. The main part
of research in social sciences was done in CNRS research centres and the
EHESS, disconnected from either the Grandes Écoles or the university system.
Universities (with some exceptions in terms of disciplines – medicine, law –
or cities) attracted neither the best students, nor the most advanced research.
300
Nevertheless, as documented in Chapter Two, some attempts were made to create

a sort of administrative science combining faculties of law and state civil service
practitioners and that had some influence in a few law faculties. But, within a
conservative environment in universities, it took a long time for some disciplines
to progressively become autonomous, especially political science. In the prestigious
law faculties, research was not prominent and social sciences were rejected.
The diffusion of policy process studies within the academic world:

from innovation at the periphery to institutionalisation in existing
disciplines
As argued in other chapters, policy studies were a latecomer in the French
academic world. On the one hand state civil servants (in fact different kinds of
them, often in competition) produced the expertise, defined policies and were in
charge of them. On the other hand, French political science did not take policy
studies as an interesting object of inquiry. Innovations came from the periphery:
from CNRS research centres in sociology and political science and from centres
outside Paris where new institutions have been created after 1945 (Institute of
Political Studies [Institut d’Études Politiques] in Bordeaux and Grenoble, then
in Strasbourg, Lyon, Aix, Toulouse, Rennes and Lille; new specialised schools
related to specific policy domains, for example the National Institute for Transport
and Safety Research [Institut national de recherche sur les transports et leur sécurité
(INRETS)] for transport policies, the National School for Public Health [École
Nationale de la Santé Publique] for health policy, and the Paris Institute for Urban
studies [Institut d’Urbanisme de Paris (IUP)] for urban policies and others). The
core institutions of French political science, Paris I-Sorbonne and Sciences Po,
showed little interest until the late 1990s.
For the sake of clarity, this part is divided into three periods. The first starts in
the late 1960s with the importation of policy analysis from the USA, mainly by
CNRS sociologists based outside French universities with links to policymakers.
The academic marginality of policy analysis at that time is a first explanation
of the division between policy analysis and policy studies. During the 1980s
an important turn occurred with the growing importance of policy studies in
political science, focused on the analysis of the policy process. The 1990s and
the beginning of the twenty-first century, the second period, corresponds to
the academicisation of policy studies with a strong development of research
programmes, publications and policy studies teaching in the existing disciplines.
The third and current period is characterised by evolutions beginning to bridge
the gap between policy studies developed in the academic world and policy
analysis in the world of policy practice.
301
The development of policy analysis at the fringes of universities (1960s–1980s)
Leca and Muller’s influential paper (2008) reviewed the three main intellectual
influences that fed the rise of policy studies in France, namely administrative
science and public law (see Chapter Two), the sociology of organisations
(developed at the CSO) and Marxism. Administrative science was a subset of
public law approaches. Marxists developed a critical approach of the state (at the
national and local level around urban policy, housing and the welfare state). The
same applies to the two leading figures of French social sciences of the 1970s:
Pierre Bourdieu and Michel Foucault. They were vigorously opposed to the
more instrumental policy analysis taught in US public policy schools. Therefore,
sociology of organisations was the sub-discipline that played the leading role in
the transfer of policy analysis from the United States. As already noted by Fabrice
Hamelin in Chapter Three, Michel Crozier played a key role in the academic
knowledge transfer between France and the United States. Crozier, close to French
modernist elites, but not a critical scholar, had been in close connection and
taught at Harvard and later went to Standford. CSO became an internationalised
research centre where its nearly official ambassador, Martha Zuber, for years
organised the visits of scores of US scholars interested in France. The connection
between CSO and US leading policy and organisation sociologist scholars has
been a major explanation of the diffusion of US policy studies scholarship in
France. Some became long-term companions to CSO such as Suzanne Berger
from MIT or Peter Hall from Harvard.
This transfer did not directly concern policy analysis, it was more focused
on sociology of organisations, especially the sociology of administration.
Notwithstanding, in the CNRS autonomous research centre created by Michel
Crozier in 1964 (the CSO) policy studies were developed, especially on local
policies (Grémion, Worms), urban policies (Thoenig) and industrial policies
(Friedberg). After Crozier, the next generation of policy studies sociologists,
such as Thoenig or Jean-Gustave Padioleau, also made the trip to the US. The
research developed at the CSO was often connected to administrative demand
and therefore integrated to public policies themselves. Michel Crozier was part
of a group of modernist civil servants the ‘club Jean Moulin’. Both the planning
body, the General Planning Commission (Commissariat Général du Plan [CGP])
and the Centre for the Coordination and Orientation of Economic and Social
Development Research (Centre de Coordination et d’Orientation des Recherches sur
le Développement Economique et Social [CORDES]) financed a long-term research
programme at the CSO including public policies and leading to publications
that became landmarks of French social science research (Bezes and Chauviere,
2004). This connection, which is the second aspect of this period, is not specific
to the CSO. At the end of the 1960s the CGP reoriented his activity on research
financing with the creation of the CORDES (see Chapter Eleven) and different
specialised research centres were created in the 1970s and early 1980s: the Centre
for Sociological Research on Law and Penal Institutions (Centre de Recherche
302
Sociologique sur le Droit et les Institutions Pénales [CESDIP]), by the Justice Ministry
and the CNRS to work on crime and justice, the centre for Cultures and Urban
Societies (Cultures et Sociétés Urbaines [CSU]), another CNRS research centre,
to work on urban policies in relation to the Ministry of Infrastructure, or the
Research Group for Social and Health Policies (Groupe de Recherches de l’Action
Sociale et Sanitaire [GRASS]) at the EHESS to work on social policies in relation
to the Social Affairs Ministry (Le Galès, 2015).
The third aspect is that these different research structures were neither located
in universities nor directly related specifically to an academic discipline. The CSO
as a CNRS research centre, was neither part of the university system and nor
was it involved in the teaching of civil servants in Grandes Écoles. Like the CSO,
other academic research institutes working on public policies have been created
outside universities: the main example is the Research Centre on Administration
and Territories (Centre d’Etudes et de Recherches sur l’Administration et le Territoire
[CERAT]) at the IEP Grenoble (see Chapters Three and Four). Teaching of
public policies was also developed in a business school: the ESCP under the
impulsion of Padioleau who was another key figure of the transfer of policy
studies from the US to France (Padioleau, 1980). However the main figure of
the diffusion of policy studies was Jean-Claude Thoenig, a trained sociologist
who started his academic career at the CSO and worked on public policies in a
policy analysis perspective. He was involved in the first European Consortium
for Political Research (ECPR) workshop on public policies in 1977 organised
by Bruno Dente, focused on organisational analysis, public administration and
local governments in a comparative perspective and aiming to produce relevant
knowledge to policymakers. After leaving the CSO, Thoenig created the Public
Policy Analysis Group (Groupe d’Analyse des Politiques Publiques [GAPP]) in the
mid-1980s, a pluri-disciplinary CNRS research institute without direct relation
to a university where a group of researchers (J. Commaille, Patrice Duran, Pierre
Lascoumes and others) developed the sociological analysis of public policies. He
was also, with Patrick Gibert, a professor of management, the founder of the
first French policy analysis journal in 1983: Politiques et Management Public. This
journal, created mainly by sociologists and management specialists, claiming its
scientific orientation (Gibert, 1983), was also trying to combine academic articles
and practice-oriented papers written by public policy actors and/or experts. This
policy analysis perspective is also obvious in the several conferences organised by
the journal (and then published) mixing different kind of participants from and
outside the academic world.
In a nutshell, the diffusion of policy studies in academic institutions started
in the 1960s, with strong links to policy practice. But it didn’t lead to the
institutionalisation of policy analysis and was located in marginal institutions of
the academic system. This situation changed with the development of policy
studies in the existing academic disciplines, especially political science.
303
The academicisation of policy studies (1990s–2000s)
As stressed in the previous section policy analysis was imported and developed in
France mainly by sociologists and only marginally by political scientists (essentially
in Grenoble with Bruno Jobert, Pierre Muller and Lucien Nizard who worked
with different policy actors and also shared policy oriented objectives). Three
explanations may be put forward: the first one is the late autonomisation of French
political science from public law (mainly in the 1970s). Therefore studies on the
state and public administration were dominated by the ‘administrative science’
perspective traditionally strongly linked to the public law approach (Leca and
Muller, 2008). Second, the autonomisation of political science from public law
was based mainly on the development of research about elections, parties, political
systems, institutions, area studies, as the core field showing the specificities of a
non-law oriented approach focused on the understanding of political behaviour.
In the 1980s the sociological approach of the state was also developed (Badie and
Birnbaum, 1979) based on the critic of the political development perspective
(Badie, 1984). However, the historical and sociological approaches endorsed by
these authors neglected the role of public policies in the process of the construction
of the state (Hassenteufel, 2007). Third, French political science was rather self-
centred. American classic political science was well known but until the 1980s
there was little interest in political science in other countries, and the development
of policy studies remained unnoticed or ignored.
Nevertheless policy analysis was not totally absent from academic political
science in the 1980s, even if in his study of the situation of political science
in France Pierre Favre mentioned the fact that policy analysis was a subfield
with very few specialists (Favre, 1982). During the first national congress of the
French Association of Political Science (Association française de science politique
[AFSP]) in 1981 a roundtable on ‘Policy Analysis’ was organised by Leca and
Jean-Louis Quermonne (both became President of the AFSP some years later).
And, in 1985, when the first French Political Science treaty (Grawitz and Leca,
1985) was published, one volume (among four) was devoted to policy analysis,
though directed by a sociologist (Thoenig). Political scientists authored seven
of ten chapters of the volume, a major breakthrough with long last influences.
This situation changed in the 1990s with a progressive development of teaching,
research, academic manifestations and publications in relation to policy studies.
Once again, innovations came from the periphery: in Grenoble most importantly
but also in Bordeaux, Montpellier or Rennes. This trend begins to be obvious
in the mid-1990s with the increasing role played in political science at the centre
(Sciences Po, Association Française de Sciences Politique) by senior scholars such as
Muller and later Lascoumes at Sciences Po, in sociology by Christine Musselin
and Philippe Urfalino at CSO, Thoenig and Duran at the École Normale Supérieure
(of Cachan). The change was also supported by the coming of age of a young
generation of comparative European public policy scholars trained in the US,
in England, Germany, Switzerland, or at the European University Institute in
304
Florence. The analysis of political science courses in law faculties (where political
science is mostly taught in French universities) in 1995–96 realised by Favre
(1998) showed that courses related to public policies (policy analysis, public
administration, local policies, European policies, social policies) represented 12.7
per cent of the total amount. Even if it was still a minor part, it corresponded to
an over 100 per cent growth compared to how it was in 1977–78, with 5.9 per
cent of the political science courses related to public policies (exclusively public
administration courses) at that time. The proportion of masters (then called
Diplômes d’Études Approfondies) in law faculties including policy analysis was even
higher: 22 per cent (9 among 41) in 1995–96.
Law faculties including political science departments were however the
rearguard. This growth of policy analysis teaching was spectacular in the IEPs
where the main CNRS research centres developing policy studies were located
at that time: especially the IEP Grenoble (CERAT), the IEP Bordeaux (Centre
d’Etudes et de Recherches sur la vie locale [Research Centre on Local Politics])and
the IEP Rennes, the Centre for Administrative and Policy Research (Centre de
Recherches Administratives et Politiques [CRAP]). This change can be related to
the development of political science in the French academic world based on the
creation, in the early 1990s, of political science bachelors and of two new IEPs
(in Rennes and Lille). It gave rise to a growing recruitment of CNRS research
fellows and assistant professors in political science.
The resistance of political science should not be underestimated, however. For
years, foreign scholars working on France had tried to convince their colleagues
from the core, that is, Paris I-Sorbonne and Sciences Po, to develop policy studies.
Berger (MIT), Dente (Bologne), Hayward (Hull and Oxford), Wright (Nuffield
College Oxford) had developed very strong connections and cooperation within
French political science, for instance with Jacques Lagroye, a dominant figure of
French political science for three decades at Paris I-Sorbonne. Although that type
of research was known it was unambiguously rejected, and hardly appeared in the
main handbook of political science written by Lagroye in the early 1990s. On
the other hand, political science at Sciences Po, until the late 1990s, developed
with little interest in public policy (with the exception of a marginal small
research centre, the Research Centre on Administrations [Centre de Recherches
Administratives, CRA]). The only exception was Yves Mény, who had spent a year
in Cornell and was professor at the European University Institute of Florence in
the 1980s. At Sciences Po, beyond failing to develop policy studies teaching for
future civil servants, he tried to influence political science and to develop public
policy, regularly inviting Dente, Hayward or Wright. It proved difficult and did
not raise much interest. He left to become the head of the Schuman Centre in
Florence. Jean Leca, who, in the 1990s, managed to organise the development
of policy studies at Sciences Po (with Quermonne) always argued that a political
science professor at Sciences Po accepted a sort of gentleman’s agreement with
French administrative elites who run Sciences Po. They were part of the governing
coalition but they were not supposed to study policies. It was only with the arrival
305
of Pierre Muller at Sciences Po, attracting students, colleagues (for example Pierre
Lascoumes) and younger scholars that research on public policies developed there
from the late 1990s onwards.
Three evolutions can be put forward in order to explain this growing position
of policy studies in French political science: the transformations of public
policies themselves, the dynamics of public policy research in France and the
reforms of the higher education system. First, French public policies traditionally
dominated by the state have been deeply transformed by two major changes.
The first one is the decentralisation process (officially starting in 1982) which
transferred new competences and new (institutional and financial) resources to
local governments. They (urban governments in particular) progressively defined,
decided and implemented more autonomous public policies (especially urban,
transport, economic and cultural policies). This strong development of local
policies triggered the development of new courses on local government and the
creation of new master’s degree courses specialising in local policies because of
the new professional opportunities at this level. It also led to the creation of a new
school specialising in the training of upper-level local civil servants (the Local
Civil Servants National Centre [Centre National de la Fonction Publique Territoriale,
CNFPT]) where some leading figures of French policies studies taught (Thoenig,
Gibert, Duran). The second change was the ever-increasing Europeanisation of
public policies leading to the completion of the single market and the adoption of
the Maastricht treaty. These two major trends have widened the scope of policy
studies, until then dominated by a strong national and statist perspective. The
AFSP organised two major conferences on these topics: the first one in 1991, ‘Is
the French model of administration in crisis?’ (‘Le modèle français d’administration
est-il en crise?’), the second in 1994, ‘Public policies in Europe’ (‘Politiques publiques
en Europe’). Both were published in the new series created by Pierre Muller at
l’Harmattan publishing house, entitled ‘Political Logic’ (‘Logiques politiques’).
Other conferences, organised by the AFSP (on the policy networks approach,
or on cognitive approaches, for example), as well as several PhDs studying policy
processes at the local, national and/or European level were published in this
important series for the diffusion of policy studies.
The development of local and European policies, deeply transforming national
policies, gave a decisive impulse to comparative policy process studies in a
political science perspective different from administrative science and taking the
institutional and political dimensions more into account than the sociological
approaches mainly inspired by the sociology of organisations (see Chapter
Four). Pierre Muller created a public policy standing group of the AFSP in
1994 strengthening the position of policy studies in the discipline. It organised
several workshops and conferences and triggered the publication of special issues
of the French political science journal, Revue française de science politique (RFSP)
focusing on public policy: five between 1992 and 2005. It was also in the 1990s,
that French handbooks were published. The first one was written by Mény and
Thoenig in 1989. Its content was mainly the presentation of the international
306
literature on policy analysis (on the different policy sequences and comparative
approaches). Then, in 1990 Muller published a shorter synthesis more focused on
his own cognitive approach and on the analysis of public policies in France; he is
also the co-author (with Yves Surel) of another handbook published afterwards
(in 1998). This academic development of policy analysis in political science was
mirrored in the 2000s by the publication of many handbooks: by Gilles Massardier
(2003), by Jean-Pierre Gaudin (2004), by Pierre Lascoumes and Patrick Le Galès
(2007), by Patrick Hassenteufel (2008) and by Jacques de Maillard with Daniel
Kübler (2009). Their content reflects the priority given to the presentation of
approaches, that is, to theoretical and methodological issues, over the description
and analysis of the transformations of specific sectors.
Third, the institutionalisation of policy studies was also facilitated by the slow
transformation of the higher education system. As documented by Musselin
(2001), the division between Grandes Écoles, university and CNRS research centres
started to diminish, an on-going process. CNRS research centres mostly became
joined research centres between either university or Grandes Écoles and the CNRS.
Those research centres would therefore comprise both CNRS researchers and
university professors. Despite many rivalries, those joined centres gave a major
boost to the development of research, including policy studies, within universities
in political science and sociology. Within CNRS, a special section was created in
the 1980s bringing together organisation and labour market sociology, and political
science. It became a stronghold of policy studies scholar very much involved in
comparative analysis. Also, this paved the way to the long-term transformation
of CNRS staff: researchers started to teach more regularly and consistently. The
teaching of policy studies expanded massively both in universities and IEPs (in
Paris and outside Paris).
Yet, this academic affirmation of public policy studies in existing disciplines (not
only in political science but also in sociology, managerial sciences and economics)
with a strong focus on theory building and discussion has widened the gap with
policy practice as the weak position of academics in the development of policy
evaluation (dominated by administration and consultants), especially sociologists
and political scientists (Lacouette-Fougères and Lascoumes, 2013, 140) and the
absence of creation of specific policy analysis schools or departments show. But
some slow changes are currently on the way.
Bridging the gap between policy studies and policy analysis? (2010s)
The trend towards the academic affirmation and institutionalisation has been
strengthened in the most recent period especially in political science were policy
studies are considered as one of the most important disciplinary subfields after
political sociology.
A good indicator is the number of candidates for the national qualification
process. This qualification is the condition to become a candidate for an assistant
professor position in a university. In the recent years the proportion turns around
307
15 per cent for public policies (16 per cent in 2011, 14 per cent in 2013, 18
per cent in 2015 [CNU, 2011–15]). Another related indicator is the number
of recruited assistant professors: between 2006 and 2012, 23.6 per cent of the
recruited assistant professors in political science had a public policy profile (Déloye,
2012, 124). And in the survey conducted in 2009 by Emiliano Grossman on
French political scientist policy studies appear as the second subfield behind
political sociology (Grossman, 2010, 569). This position of policy studies in
political science is also reflected by the number of roundtables on public policies
at the French Political Science Congress (around 25 per cent of the selected panels
for the 2017 Congress deal with policy studies) or the presence of public policy
specialists at the political science section of the National University Council
(Conseil national des universités [CNU]), in the national jury recruiting full-time
professors every two years, in the fortieth section of the national commission of
the CNRS and in the editorial board of the French political science journal, RFSP.
Significantly, a new journal focused on public policy and public administration
(Gouvernement et action publique) was created in 2012 and has the same publisher
as RFSP: the Presses de Sciences Po. It is a more academic journal than Politiques
et Management Public, focused on political science and sociology (excluding the
more practice oriented managerial sciences), with an internationalised editorial
board. Its content reflects the increasing links between policy studies and other
areas of political science (political sociology, international relations, European
studies, comparative politics and so on) and sociology (sociology of administration,
sociology of professions, sociology of science, for example) so as the growing
internationalisation of French policy studies. A growing number of scholars
are involved in international associations (especially the International Political
Science Association (IPSA), with the creation of the International Public Policy
Association (IPPA) by Philippe Zittoun, the European Consortium for Political
Research (ECPR), and the Council for European Studies), publish in English,
develop comparative policy studies programmes, and are involved in international
public policy journals, for example.
Internationalisation is also considered as a main factor for the development of
gender perspectives in French policy studies. Starting in the 2000s, transnational
connections with the feminist comparative policy community have nurtured
researches on gender policy studies made by French scholars through various
European projects and on comparative studies of gender equality policies (Mazur
and Revillard, 2016, 563). Inside the French political science association Pierre
Muller, at that time general secretary, played a pioneering role during the
mid-2000s (Engeli and Perrier, 2015, 355) by organising workshops and an
international conference on gender and public policy (then published in 2008:
Engeli et al). In 2009 a special issue of the RFSP on gender between policy
and politics was published (Boussaguet and Jacquot, 2009) and since that period
several scholars working on the gender dimensions of public policies have been
recruited in French universities and research centres. The gender dimension is
now systematically integrated into public policy researches and courses, as a part
308
of a more general trend towards the growing inclusion of gendered approaches

in political science and sociology.2
If the academic position of policy studies has been strengthened (especially
for political science and to a lesser extent in sociology), the development of
policy process studies has also led to a dilution (in relation to the adoption of a
political sociology perspective) and fragmentation of policy studies (in relation
to a specialisation in policy domains: for example, urban policies, justice and
crime, welfare, health and the environment). Nevertheless, this fragmentation and
dilution is also a sign of healthy intellectual debates between different approaches
and a renewal of methods.
It is also important to note that the French public administration is increasingly
interdependent with international organisations and national administrations all
over Europe and the EU. The splendid isolation doesn’t work anymore, and the
French state is, like others, being reconfigurated (King and Le Galès, 2017). It was
quite a shock for top French civil servants to discover that they were now turned
down for jobs at the World Bank or the OECD because they did not have a PhD.
Progressively some sectors of the French Administration are understanding the
necessity to work with academic policy studies. The Council of State now works
with law university professors and occasionally might mobilise policy studies. In
negotiating in Europe or elsewhere, the Ministry for foreign affairs has learned
to work with academics including policy studies in a systematic way. With the
decentralisation process, many local and regional politicians have developed
systematic interactions with policy studies research, especially at the regional
level. The French state has created institutes, developing executive education
programmes where policy studies play a major role, in several ministries. Executive
education and policy evaluation is on the rise, including for civil servants, and
policy studies experts are involved in those flourishing programmes as well as in
the new expertise institutions created in some policy fields since the end of the
1990s. Even Sciences Po has finally opened a public policy school (2015) where
future candidates for ENA are now obliged to be trained in economics, in policy
process studies and sociology. Furthermore, some Grandes Écoles engaged in a
long-term transformation to be closer to an international model. Sciences Po
paved the way to become officially (in 2015) an international research university,
still elitist but comprising not 2,000 but 13,000 students with a steady increase
of the number of full-time academics in the teaching and the organisation
combining research, teaching and close connections to professionals. This
trend also concerns business schools, the École Normale Supérieure and Engineer
Grandes Écoles. A growing number of Grandes Écoles have now a five-year
curriculum, deliver master’s degrees and even PhDs. This is also related to the
most recent development of French higher education policy with the creation
of large communities of higher education institutions, Communauté d’universités
et établissements (COMUE) composed of universities, Grandes Écoles and CNRS
research centres, in which new master and doctoral programmes specialising in
309
public policies can be developed, blurring the traditional differentiation between

professional and research masters.
Conclusion
This chapter stressed two main factors of the gap between policy analysis and
the academic world in France: the first one is the core role of the Grandes Écoles,
located outside universities and separated from research, in the training of future
policy elites; the second one is the strength of disciplinary logics in the academic
field, contradicting with the pluri-disciplinarity required by policy analysis. It
helps to understand why policy studies started, and have been developed, in
institutions more opened to various disciplines (and historically located at the
fringes of the academic system): the CNRS and the Institut d’Études Politiques,
as has the role of political science. This more recent (and weaker) discipline, at
the crossroads of different social sciences, has been more welcoming than others
for the analysis of public policies. Notwithstanding policy studies have been
developed in already existing disciplines, without creating neither a new one,
nor new specific academic institutions.
Nevertheless, the days of the splendid isolation between elites from the Grandes
Écoles running the French state and academic research may be (nearly) gone. In
this slow, but systematic, long-term transformation, more connections between
policy studies and policymakers can be noticed (with the growing participation
of academics in policy expertise forums) and the trend towards the creation of
French public policy schools is on the way. Policy studies are on the rise not only
in universities but also in some Grandes Écoles, they are now well established and
more integrated and recognised in international research circuits. The mixture
of influence of sociology, political science, political economy and ethnography
provokes many debates and interesting innovations in terms of methods and
research questions, even if the tensions, on the one side, between disciplines,
especially with the strengthening of economic policy analysis and economic
policy evaluation (see the previous chapter), and on the other side, between
disciplinary logics and the pluri-disciplinarity of policy analysis, are far from
having disappeared!
Notes
1 The website of the Conférence des Grandes Écoles gives a list of 222 schools.
2 In relation with the feminisation of French social sciences: between 2005 and 2015 the number
of women among assistant professors has increased from 32 per cent to 42 per cent in political
science and from 42 per cent to 54 per cent in sociology; for full professors it has increased from
10 per cent to 20 per cent in political science and from 20 per cent to 33 per cent in sociology
(data taken from Faudot, 2017).
310
References
Badie, B. (1984) Le développement politique, Paris: Economica.
Badie, B., Birnbaum, P. (1979) Sociologie de l’État, Paris: Calmann-Lévy.
Bezes, P., Chauviere, M. (eds) (2004) L’État à l’épreuve des Sciences sociales, Paris:
La découverte.
Bourdieu, P. (1989) La noblesse d’État, Paris: Minuit.
Boussaguet, L., Jacquot, S. (eds) (2009) ‘Le genre entre policy et politics’, Revue
française de science politique, Special Issue, 59(2), 173–351.
Charle, C., Verger, D. (2007) Histoire des universités, Paris: Presses universitaires
de France.
CNU (Conseil national des universités) (2011–15) Rapport annuel du Conseil
National des Universités, section 04 (science politique), Political Science Section,
Paris: CNU.
Déloye, Y. (2012) ‘Eléments pour une morphologie des politistes français au vingt
et unième siècle’, Politique et Sociétés 31(3), 109–26.
Dobuzinskis, L., Laycock, D.H., Howlett, M. (2007) Policy Analysis in Canada:
The State of the Art, Toronto: University of Toronto Press.
‘French touch’ dans l’analyse des politiques publiques?, pp. 349–75, Paris: Presses de
Sciences Po.
Engeli, I., Ballmer-Cao, T.-H., Muller, P. (eds) (2008) Les politiques du genre,
Logiques politiques, Paris: L’Harmattan.
Faudot, D. (2017) Les carrières des enseignantes-chercheuses dans l’enseignement supérieur,
Paris: Syndicat national de l’enseignement supérieur, Fédération syndicale
unitaire (SNESUP-FSU), www.snesup.fr/sites/default/files/fichier/femmes_et_
esr.pdf?ct=t(La_Lettre_du_Sup_Lettre_d_information_du_5_30_2017)&mc_
cid=c5c9e75eb6&mc_eid=a13869aff6.
Favre, P. (1981) ‘Les sciences de l’État entre déterminisme et libéralisme: Emile
Boutmy et la création de l’École Libre des Sciences Politiques’, Revue Française
de Sociologie 22(3), 429–65.
Favre, P. (1982) ‘La science politique en France depuis 1945’, International Political
Science Review 2(1), 95–120. (Translation of ‘Political science in France’, in W.
Andrews (ed) (1982) International Handbook of Political Science, pp. 154–68,
Westport, CT: Greenwood Press.)
Favre, P., Legavre, J.-B., (eds) (1998) Enseigner la science politique, Paris: L’Harmattan,
Gaudin, J.-P. (2004) L’action publique, Paris: Presses de Sciences Po.
Gibert, P. (1983) ‘Éditorial’, Politiques et management publics 1(1), 1–3.
Grawitz, M., Leca. J. (eds) (1985) Traité de science politique, Paris: Presses
Grossman, E. (2010) ‘Les stratégies de publication des politistes français’, Revue
311
Hassenteufel, P. (2007) ‘L’État mis à nu par les politiques publiques?’, in B. Badie,

Y. Déloye (eds) Les temps de l’État: Mélanges en l’honneur de Pierre Birnbaum, pp.
311–29, Paris: Fayard.
Heilbron, J. (2015) French Sociology, Ithaca, NY: Cornell University Press.
King, D., Le Galès, P. (eds) (2017) Reconfigurating European states in crisis, Oxford:
Lacouette-Fougères, C., Lascoumes, P. (2013) ‘“Les scènes multiples de
l’évaluation”: Les problèmes récurrents de son institutionnalisation’, Laboratoire
interdisciplinaire d’évaluation des politiques publiques (LIEPP) Policy Paper 1, Paris:
LIEPP-Sciences Po.
Colin.
Le Galès P. (2015) ‘Les approches françaises de politique publique au temps de
Pierre Muller: Complément d’enquête’, in L. Boussaguet, S. Jacquot, P. Ravinet
(eds) Une ‘French touch’ dans l’analyse des politiques publiques?, pp.185–201, Paris:
Presses de Sciences Po.
en France’, in O. Giraud, P. Warin (eds) Politiques publiques et démocratie, pp. 185–
99, Paris: La Découverte.
Maillard, J. de, Kübler, D. (2009) Analyser les politiques publiques, Grenoble: Presses
universitaires de Grenoble.
Massardier, G. (2003) Politiques et action publiques, Paris: Armand Colin.
Mazur, A.G., Revillard, A. (2016) ‘Gender policy studies: Distinct but making
the comparative connection’, in R. Elgie, E. Grossman, A. Mazur (eds) The
Oxford Handbook of French Politics, pp. 556–82, Oxford: Oxford University Press.
Mendras, H. with Le Galès, P., Oberti, M. (1995) Comment devenir sociologue:
Souvenirs d’un vieux mandarin, Arles : Actes Sud.
de France.
Muller, P. (1990) Les politiques publiques, Paris: Presses universitaires de France.
Muller, P., Surel, Y. (1998) L’analyse des politiques publiques, Paris: Librairie Générale
de Droit et de Jurisprudence (LGDJ).
Muller, P., Quermonne, J.-L. (eds) (1995) Les Politiques publiques en Europe, Paris:
L’Harmattan.
Musselin, C. (2001) La longue marche des universités françaises, Paris: Presses
Padioleau, J.-G. (1980) L’État au concret, Paris: Presses universitaires de France.
Thoenig, J.-C. (ed) (1985) Les Politiques publiques, vol 4 of M. Grawitz, J. Leca
(eds) Traité de science politique, Paris: Presses universitaires de France.
312
EIGHTEEN
Public policy analysis in France: from public

action to political power
Patrice Duran
Compared to other countries in Europe, public policy analysis in French

universities began rather late. Indeed, this research approach proved largely
dependent on the state in which it was undertaken and on the nature of the
elites that were likely to drive it. In addition, it depended on cultural elements
that shaped how public affairs were defined and justified. In this regard, France
subscribed to a highly structured unitary state in which legal reasoning not only
played a historically dominant role in the training of civil servants but also dictated
how they carried out their duties (Duran, 1990). Moreover, the very idea of public
policy had little empirical clout and even less scope because the technical reasons
associated with public finance made it impossible to adequately connect policy to
specific French programmes that might have been granted specific funds within the
budget. It was not until the reform of the legal framework of the finance laws of
1 August 2001, upon the adoption of the organic law relating to the finance laws
(LOLF), that public spending began to be correlated with public policy objectives
in order to present state policies as ‘performance programmes’. It is noteworthy
that Raymond Aron, one of the few French authors to evoke the term ‘policy’
in the 1950s and 1960s, spoke only, and rather vaguely, of programmes of action
through which individuals and groups could assert their interests (Aron, 1962;
1965). Given the absence of more in-depth analyses, references to ‘policies’ and
‘programmes’ remained ambiguous from both a theoretical and an empirical
perspective. These terms were associated with ‘fields’ as vague as social, cultural,
economic and industrial policies. In no way was policy analysis associated with
a genuine research approach.
Extending beyond the inevitable fashion trends around the discovery of a new
approach, however, policy analysis established itself from the 1980s. It emerged as
indispensable to the study of political power and soon became a dominant field
of research within French political science. Even today, policy analysis is fully
acknowledged by other social science disciplines such as sociology, economics
and even law (Caillosse, 2016; Mockle, 2007). The reception and development
of public policy analysis in France is strongly tied to the question of the state for
several reasons. First, the state is perceived as central in French political culture
because it played an important role in the political integration of the country
and because it embodies a powerfully structured state highly differentiated from
civil society. Second, public policy debate was initiated at a time when political
sociology had just ‘rediscovered’ the state following, in particular, the closure of a
313
major socio-historical research programme supervised by Charles Tilly about the

formation of European states. This ‘rediscovery’ led to the creation of a genuine
‘sociology of the state’ which represented a major scientific challenge for the study
of political power (Tilly, 1975; Badie and Birnbaum, 1979; Leca 1980). Third,
the discovery of policy analysis occurred alongside a crucial and unique moment,
that is, during the decentralisation reform in 1982. This reform prompted debate
on the actual role of central government in the management of public affairs. In
sum, French public policy analysis could only integrate the sociology of the state,
and as a result, could do no more than revisit and complement it.
This explains French researchers’ interest in this new analytical approach. Policy
analysis henceforth appeared as a powerful tool with which to analyse the state and,
more broadly, political power. Indeed, as Ted Lowi (1970) rightly argued, the state
is not a phenomenon that can be studied directly. Starting from the observation
of actions and, thus, actors, policy analysis is essentially the principle of reality of
a state ultimately defined by its actions, rather than by its roles and structures.
Policy analysis has indeed emerged as the ‘missing link’ making it possible to
associate the management of public affairs with the institutional framework that
makes this management possible (Duran, 2004). Reflecting on action calls to
attention the fact that while action is always based on intentions and capabilities,
it is also linked to operational conditions; the main characteristic of action is the
opportunity it affords us to account for the linkages between these dimensions.
Public policy analysis was attractive as it made it possible to more accurately grasp
the scope of the institutional arrangements specific to each state by analysing their
impact on the definition and implementation of public policies (Ashford, 1978).
Subsequently, in the French context, policy analysis made it possible to highlight
the impact of the changes observed in intergovernmental relations with regard to
the management of public affairs. Public policy analysis thus offered a remarkable
tool with which to interpret political power by integrating its constituent
dimensions, that is, policy, politics and polity. It is no coincidence that modern
states were thereafter defined as Policy states. These states were characterised
by governing modes largely shaped by public policies and by multiple forms of
privatisation and delegation closely linked to the exercise of state authority (King
and Le Galès, 2011). Beyond the issue of the state, which has remained a decisive
factor, many French researchers have adopted an analytical approach that avoids
isolating policy analysis from a more general reflection on political power, that
is, on its nature, context and scope. This combination enables policy analysis to
produce the most useful insights, and it is to this approach that French researchers
have remained highly committed (Duran, 1990; Leca, 2012; Bezes and Pierru,
2012; King and Le Galès, 2011). Thus specified, public policy analysis – at least
in France – is undoubtedly the modern theory of political power.
Public policy can no longer refer to the simple characterisation of the
intervention of public authorities. Above and beyond their specific advantages,
the legal theories of public authorities or of the civil service that have largely
shaped French public law and structured the practices of civil servants do not
314
From public action to political power
allow us to fully grasp the reality and specificity of these theories in all their
aspects. The significant development of policy analysis and, more broadly, the
strong focus on the modalities of development of public action have renewed
our understanding of how public affairs are managed. Analysing policy from the
perspective of political power thus requires analysts to seriously take into account
the intervention of public authorities relative to their ability to construct and
implement public policies. This analysis provides a definition of political power
that is now complete: the ability to define collective goals, mobilise the necessary resources
in their pursuit, make the necessary decisions to obtain them and accept the outcomes.
Policy analysis leads to a strictly empirical conception of political power which
conclusively associates the reality of power with the knowledge of that reality.
What does public policy analysis really mean? What has been the impact of the concepts
it has developed to explain political reality? There is a need to draw attention to
these essential questions both in France and elsewhere. In other words, there is
a need to (1) better understand the process of appropriation that contributed to
the full development of public policy analysis on the French academic scene,
particularly from the perspective of the sociology of public action that quickly
established itself as the dominant analytical approach, and (2) analyse the social
utility of policy analysis at a time when questions about how modern societies
are managed have become increasingly common.
Public policy and public action: the singular French approach of

the sociology of public action
Today, it is widely accepted that public policy analysis is neither a ‘discipline’
nor a fully integrated theory of public action; rather, public policy analysis is
more a project and an area of research. The French case is not original in this
respect. The social sciences are based on a methodological pluralism driven by
the different ways these sciences analyse research objects. Moreover, interest in an
empirical object does not necessarily mean that the object will be automatically
transformed into a research object, especially since, as we mentioned previously,
the concepts of ‘programme’ and ‘public policy’ long remained foreign to the
French vocabulary of the management of public affairs.
The appropriation of policy analysis became a preoccupation of French research
and, in both sociology and political science, was reflected in particular by the
progressive shift from the concept of public policy to that of public action: from
the analysis of policies strictly speaking to a political sociology of public action. The
preference for the concept of public action significantly contributed to the
development of a research agenda specific to French research which became
increasingly autonomous with time.
The public action concept emerged in the 1990s following critical reflection
around the dissatisfaction with how the term ‘public policy’ was commonly used
(Duran and Thoenig, 1996; Thoenig, 1998; Duran, 2009). The criticism levelled
against this term was wide-ranging: its trivialised use was challenged and the
315
manner in which it was used was criticised for its inappropriate assimilation to a
rather institutionalist logic of intervention often associated with a questionable
methodological approach. Its use was also criticised for its limited analytical
scope as it embodied a type of public action that was too historically situated
(Hassenteufel, 2011).
Indeed, it seemed preferable to speak of public action rather than public policy
because, while the concept of public policy clearly underscored the voluntary
nature of action, the frequently rhetorical use of the term also promoted a
reassuring image of a government aware of what it was doing and of where
it was going. Owing to a government culture strongly influenced by the legal
theory of public authority and that of the public service, such a representation
was clearly favoured in France. It thus became necessary to liberate the country
from a concept of policy which included vague definitions and which tended to
be too easily identified only with the objectives formally pursued by government.
It was thus necessary to distance policy analysis from the basic category invented
by the policymakers themselves. This ‘disenchantment of politics’ driven by the
sociology of public action sought to bring attention to the fact that not all public
actions necessarily embody public policy, nor are they necessarily part of policies
or action programmes.
Defining action as an intentional act means that we must define public policy
as a systematic intention and, unless the expression is to be rendered completely
meaningless, accumulating disparate measures and diverse gestures does not reflect
controlled action. However, while the role of public policy is defined by specific
issues, public policies also require actors to identify sequences of actions and
connect these sequences together around specific policy objectives. The concept
of public policy cannot, in itself, symbolise all public action (Duran, 2003).
In addition, the very idea of public policy encompassed a form of public
intervention methodologically driven by a frequently top-down interpretation.
This was influenced by an essentially state-centred analysis of public decision-
making, even as social sciences were attempting to rectify this view by introducing
bottom-up approaches. Bottom-up approaches were an attempt to incorporate
the most recent contributions in the sociology of action into public policy
analysis (Thoenig, 1976; Elmore, 1979; Hjern and Porter, 1981; Sabatier, 1986).
While there seemed to be a general consensus about the importance of taking
into account the implementation phase during the policy-making process, the
corresponding analysis revealed highly varying research approaches. Methodology
posed the greatest challenge. It was less an issue of a specific type of intervention
than the criticism of a method of analysis far too dependent on the established
frameworks of public policy, and one that paid too little attention to actors’
interactions within concrete systems of action, interorganisational networks or
policy subsystems. This reflected an image that was somewhat counterintuitive
to a fragmented political system largely characterised by the diversity of problems
that needed to be addressed and by the poor coordination of corresponding social
systems; the situation proved conducive to alternative modes of reasoning. The
316
increased focus on public issues would also prove decisive; analysis helped highlight
the change in nature as well as the complexity of these issues.
Indeed, it has become increasingly clear that the approaches used to address
public issues today go beyond the scope of government institutions and do not
always take the form of clearly institutionalised policies. The identification of the
interactive relationships and numerous interdependences that span institutional
frameworks challenges the most common boundaries between the public and
private spheres and between levels of government, and has thus led to fresh
questions about the issue of political power. The concepts of governance and
multilevel governance with regard to the specifically European dimension of
public action have been the most representative indicators of this new focus.
Naturally, in the attempt to grasp the reality of political power in a multipolar
and pluralistic world, a few difficulties were encountered. The unclear points of
reference essentially resulted from the gap between the representations that still
largely shaped how France interpreted politics and the experience of a public
action that did not fit within those mental frameworks. The state remained a
central issue. Multiple studies about the concept of ‘governance’ described the
emergence of a new age of public action characterised by the complexity of
public issues, increasing interdependence and international competition, the
existence of multiple decision-making centres lacking clear hierarchy and the
development of informal decision-making methods to offset the lack of flexibility
in bureaucratic structures. Owing to both European integration and to the
processes of decentralisation that were being developed in all member states and
in France in particular, ‘polycentric forms of power’ emerged. Moreover, there
were internal changes in a public action that increasingly distanced itself from
the logic of production characteristic of the formation of a welfare state and
became increasingly characterised by ‘the fluidity of public issues, the uncertainty
of territories and the fragmentation of political power’ (Duran and Thoenig,
1996: 582).
Ever since, the term ‘public action’ has gained the upper hand. As Pierre
Lascoumes and Patrick Le Galès accurately describe, ‘the notion [has been]
widely adopted by most contemporary writers’ (Lascoumes and Le Galès, 2007:
17). This systematic use has gradually been taken into account to develop the
political sociology of public action. Although this perception of public action
is multifaceted, it has shifted away from a policy analysis considered somewhat
‘orthodox’ (Muller, 2000). This research approach has proved valuable by
producing high-quality studies around the analysis of public issues in particular,
and by developing a new and exciting sociology of public action based on
instruments (Halpern et al, 2014). Shaped by the dual influence of Weber and
Foucault and the combination of several sociological approaches including the
Sociology of Science and Technology, current French thought on the instruments
of government has enriched and built on a decisive issue – although this line of
thought remains indebted to the pioneering work of Christopher Hood. French
thought has undoubtedly revived the issue of policy implementation. Moreover,
317
it has introduced substantial changes in the approach by proposing a different type

of reasoning which has revealed that the development of management systems
involves far more than mere actors’ games. Developing a system of governance is
not a trivial affair. Beyond the effects produced, this development also highlights
the effort to articulate the making of public policies with the distinctiveness of
how government structures have been built in order to reduce any decoupling
between the construction of administrative organisations and public policy-
making; this is a decisive issue as we shall see below.
By restoring the link between policy implementation and policy outcomes,
the sociology of instruments clearly revives what, in my opinion, lies at the
heart of the debate initiated by public policy analysis. The strength of the mode
of reasoning in terms of public policy within what is now commonly referred
to as public policy analysis has arisen from how it connects problems, choices,
means, outputs and outcomes. Indeed, reflecting on public policy means putting
into perspective political choices, the concrete processes through which these
choices are implemented, and the ensuing outputs and outcomes. The link
between outputs and outcomes has historically defined the originality, novelty
and importance of public policy analysis. The implementation of public action
is a process whose nature is instrumental; its ability to effectively contribute to
addressing the problem that inspired its formation should be assessed.
A decision is valuable only if it is implemented; this explains why the issue of
the instruments of government is appealing and relevant. The tools chosen are
a good indicator of governance ‘styles’ and of a genuine desire to undertake an
action plan. As Hood has noted, the choice of policy instruments may lead to an
extremely heated political debate, a debate that may be even stormier than one on
the expected outcomes, simply because a policy instrument has direct and practical
implications for policy effectiveness and efficiency. Policy instruments provide
the link between what is desired and what is actually achieved. The sociology
of instruments draws its strength from the relationship it re-establishes between
outputs and outcomes. At a time when performance matters, it draws attention
to the fact that the performance of public action is defined by its outcomes. As
Peter John forcefully argues, ‘Policy outcomes are what decision-makers should be
concentrating on and responding to’ (John, 2011). Classical authors as distinct as
Weber and Dewey asserted that politicians should be held accountable for policy
outcomes; indeed, they argued that the main responsibility of politicians was to
ensure outcomes. As developed in France, the issue of instruments is consistent
with this central view. Regardless of its significance, public policy analysis cannot
simply be reduced to the construction of problems without questioning the
effectiveness of policy choices and, thus, policy outcomes. While sociologists
and political scientists abandoned the issue of outcomes, economists became very
keen on this issue. Today, economists continue to enjoy a monopoly in this area,
to the extent that French policy evaluation studies primarily mobilise the know-
how of economists, even as the latter have long distanced themselves from policy
studies (Duran, 2010). The sociology of public policy instruments à la française
318
is likely to rectify this situation and provide valuable insights into the analysis of
outcomes by reiterating the question of the ability of policymakers to produce
outcomes. This also provides an opportunity to reintroduce a causal relationship
into the analysis; when it relates to action, this is all the more essential insofar as
taking action requires explanatory knowledge.
Public policy analysis should not, however, lead the concept of public policy
into oblivion; indeed, abandoning this concept is not without its risks. Reflecting
on public action has the potential to provide a general and decisive framework.
This framework makes it possible to incorporate the concept of public policy
as a form of action and to more adequately define its analytical dimensions.
Criticism has generally been levelled against the inaccurate and ambiguous use of
the term ‘public policy’, rather than against the concept in the strictest sense of
the term. Moreover, it must be said that literature has failed to provide a precise
and in-depth definition of this concept due to an approach that is more often
descriptive rather than analytical and which overlooks the fact that ‘a concept is
not knowledge, but an instrument of knowledge’, as Weber rightly stated. By
observing the intricacies of an indeterminate public action, the notion of public
policy may be lost. Undoubtedly, the usefulness of the concept of public action
lies in the fact that it is a generic concept that can be defined as all social processes
through which problems considered to fall under the responsibility of public authorities
are addressed. Conversely, public policy refers to a very specific aspect of public
action that can be defined as an action consciously directed toward ends, means
and outcomes.
A useful approach improves our understanding of social phenomena and
ultimately acts as a tool for action. Borrowing from the categories developed by
Gilles-Gaston Granger, ‘epistemic rationality’ (the rationale behind description
and explanation) and ‘practical rationality’ (the rationale behind action) differ
in nature and scope but are proving to be increasingly interdependent today
(Granger, 1995). While there is a need to distinguish the explanatory power of
the social sciences from their uses, it must be pointed out, however, that these uses
are made possible through the advancement of science. The reflection initiated
by public policy analysis stimulated the debate on evaluation, that is, on the
informed assessment of actions defined and implemented by public authorities in
an attempt to contribute to the resolution of public issues identified as requiring
their intervention. This assessment is based on an in-depth knowledge of potential
outcomes and the processes that may produce them. In terms of management of
public affairs, assessment merely reflects the imperatives of public policy analysis.
Unfortunately, public authorities have failed to take full ownership of public policy
analysis. This has frequently hampered the development of a research approach
appropriate for evaluation within the social sciences and has accorded a dominant
position to economics. Moreover, the multidisciplinary approach at the heart
of policy sciences and public policy analysis in North America has never caught
on in France. This can be explained by the fact that the institutionalisation of
disciplines in France is probably stronger than elsewhere and, instead of combining
319
the different analytical perspectives, it has pitted them against each other; this also
largely explains the absence of real ‘Schools of Public Policy’.
At stake is the shift from the analysis of policy to the analysis for policy. While
social sciences have become essential for the management of public affairs,
their role has remained largely dependent on how they are organised, which
determines the effectiveness of their joint mobilisation. In this regard, the slow
pace of learning is clearly regrettable.
Analysing political power through the prism of the theory of

public action
Reasoning in terms of public problems leads to a radical transformation of the
management of public affairs. The significant development of public policy
analysis and, more broadly, the greater focus on the modalities of development of
public policy methods have led to an increased understanding of the management
of public affairs. Moreover, it has made it possible to understand how changes
in public action have put a strain on the nature, context and scope of political
power. Put differently, in the French context, these changes have led to increased
tension in the public service, the state and in democracy.
Rethinking the public service
Public authorities have clearly lost the ownership of public problems. As in other
parts of the world, the French government had to focus on this thorny issue which
resulted from the encounter between a necessarily collective definition of the issues
to be addressed and their fragmented implementation, horizontally and vertically,
between both public agencies and government levels. Involving the authorities in
policies beyond their scope has a significant impact on their functioning. These
authorities, however, are yet to take all measures and implications into account,
which clearly explains their slow and difficult modernisation. Fundamentally, all
management styles and the whole management system must be turned upside
down in the face of the need to advance toward the desegmentation of public interventions
and ensure their coherence. Many issues call for greater flexibility in administrative
organisation and administrative action: the existence of multiple partners, the often
contingent nature of the coalitions to set up and the dynamic nature of public
issues and their permeable boundaries. Currently, integration costs are the most
sensitive points to be accounted for in the organisation of work. It is less about
mechanically implementing competencies and more about connecting different
competencies depending on the issues to be addressed; these issues increasingly
cut across official nomenclature and organisation charts. In France, the difficulty
associated with introducing administrative reforms is a reflection of the difficulty
encountered in developing a joint plan of action; this difficulty goes beyond the
issue of human and financial resources.
320
There have been successive reforms over the last 20 years and these have been
primarily characterised by intense organisational restructuring. The reforms have
generally been marked by the revision of organisation charts primarily through
decentralisation, ‘agencification’, privatisation and the merging of services. It is
still impossible today to make an overall assessment of all the reforms undertaken.
However, it may be noted that organisations, rather than more flexible forms of
coordination, are still perceived as the most convenient channels through which
to establish coordination. There is an obvious preference for tightly coupled
systems over loosely coupled systems (Landau, 1991). As a result, a paradoxical
situation has emerged: on the one hand, strong criticism has been levelled
against bureaucracies considered incapable of efficiently managing problems
judged to be way beyond their scope. On the other hand, attempts to rethink
these bureaucracies have presented them as indispensable (Bézès, 2007; Le Galès
and Scott, 2008). The proliferation of diverse agencies that emerge alongside
existing administrative structures could thus represent a solution to the difficult
integration of the sectoral logics that may be in competition in any given domain
of intervention, for instance in environmental or health issues. Finally, although it
was said that the reforms’ objectives were to ensure greater operational capability,
greater efficiency and greater effectiveness, it is clear that most of these reforms
were primarily motivated by an attempt to reduce state spending.
Indeed, the distinction between outputs and outcomes, which lies at the core
of public policy analysis, has proved to be highly destabilising for administrative
organisations in a state that has consistently developed the ‘cult of Administration
with a capital A’, to the extent of making this distinction the ‘historical treasure
of the State in France’ (Legendre, 1992). It has actually led to the emergence of
a more complex governing approach that has greatly undermined the French
government in the sense that one can now say that a public policy perspective
aims at substituting a logic of outcomes by a logic of outputs. Indeed, government
authorities are no longer mere service providers responding to more or less
demanding social issues. These authorities must now incorporate the imperatives
of the management of ‘public issues’ although they rarely have a monopoly in
addressing these issues. Moreover, the manner in which they address public issues
generates outcomes which define the reality of their performance in terms of the
nature, scope and the extent of their responsibility. The notion of outputs prevalent
in state authorities must now incorporate the notion of outcomes prevalent in public policy.
Meanwhile, assessing the activities undertaken by the authorities does not simply
involve an internal technical assessment of the quality of production. Rather, it also
involves assessing how these activities contribute to addressing situations defined
as problematic. Evaluation has now become external, which explains why an
evaluation of public policies that goes beyond the difficulties associated with their
implementation has gained in popularity; this also explains why the concerned
authorities have been highly reluctant to undertake this evaluation. Indeed,
these authorities soon discovered that the quality of their services did not justify
the legitimacy of their actions and that product quality no longer determined
321
governance effectiveness: constructing roads does not automatically translate into

land use planning, just as obtaining a degree does not necessarily mean one will
find a job. State agencies as public bureaucracies are driven by production and
are thus more sensitive to the more familiar management vocabulary than to the
more distant language of outcomes, especially as actors are generally assessed
depending on what they do rather than on what their actions generate. It is
clear that focusing on the activities of government authorities alone without
taking into account what these activities really mean can lead to no more than
an evaluation of the dimensions of production and the services offered. This can
only encourage highly managerial reflexes which have now shifted toward the
quest for savings, to the detriment of genuine cost management, as evidenced by
the so-called General Review of Public Policies (Révision Générale des Politiques
Publiques [RGPP]) policy conducted between 2007 and 2012 which had little to
do with public policy analysis. The influence of new public management (NPM)
has often contributed to the development of highly managerial approaches in
which the goals of efficiency and cost management have largely prevailed over
the analysis of outcomes which is always more difficult to establish and is also
more politically disturbing. Indeed, one of the lessons learnt from public policy
analysis in relation to paving the way for a fundamental distinction is that the
management of bureaucracy is not synonymous with policy management; the
effectiveness of one does not necessarily translate into the effectiveness of the
other. This has been a major source of tension insofar as it is easier to master
the management of public administrations than public policy, especially when
efficiency tends to outweigh efficacy and the quest for ‘best practices’ prevails over
the quality of outcomes. Public policy analysis places emphasis on the fact that
government authorities are simply an instrument for the management of public
affairs. As we previously mentioned, this is one of the potential advantages of
the debate on the instruments of government, and in particular, management
instruments (Gibert, 2010), as this debate provides a central view that makes it
easier to reflect on the relationship between organisation and policy, and thus,
more effectively to connect outputs and outcomes (Halpern et al, 2014).
From the perspective of knowledge, the success of public policy analysis led to
the loss of interest in the analysis of state agencies which, until then, had been
traditionally strong in France. The challenge today lies in better understanding
the relationship between organisational structure and the outcomes of a given
action. Clearly, there is room for a genuine research programme and French
authors have now shown great interest in this programme by focusing on the
analysis of the ‘organisational policy’ implemented by public authorities (Bezes
and Le Lidec, 2016a; 2016b). To date, French authorities have yet to develop a
genuinely collective culture of outcomes based on outcomes management. The
lack of interest in outcomes has historically led to a negative consequence, that
is, France has experienced delays in establishing reliable databases and, more
broadly, public bureaucracies have generally poor knowledge of their contexts
for action. The mobilisation of data, however, is crucial for the effectiveness of
322
public policies and for accurately assessing the impact of public programmes.
The slow diffusion of the evaluation of public policy in France is largely linked
to this unfortunate situation (Duran, 2010).
Reflecting on outcomes also results in the emergence of a different relationship
with the territory. Indeed, the territorial principle which guides public policy
debate seeks to define a dynamic space of the management of public affairs
characterised by diverse situations, potential solutions and the variability of the
territories’ geographical influence; policy implementation requires adequate
knowledge of territories. Where a detailed knowledge of the areas concerned is
lacking, there can be no quest for efficiency. Governance begins and ends by questioning
territories. Naturally, all problems are rooted in a territory. It is known, however,
that territorial scales can vary considerably depending on the nature of the issues
involved. It is not simply about action; it is also a decisive issue owing to the
effects of knowledge that have become increasingly crucial, and the considerable
increase in the distance between the spaces where decision-making takes place
and the spaces where the practical effects occur. The consequences of this new
approach to governance have been far from harmless. Because territories are
complex and always specific configurations of issues and actors, they are not
adapted to excessively automatic management methods that rarely fall under the
responsibility of a single level of government. Once again, public policy debate
has broken one of the old rules of the French legal culture of governance by
putting an end to standardised management solutions in France.
The territorialisation of public action primarily involves knowing and
understanding situations. Its vast scope includes expert knowledge, policy
instruments, renewed methods of coordinating the action and cooperation
of actors, and the institutional frameworks in which public action takes place
(Duran, 2011).
Redefining the rules of the game
Naturally, we must also question the validity, relevance and legitimacy of the
institutional framework within which public action unfolds. As we mentioned
earlier, the management of public affairs increasingly involves complex structures
comprising multiple issues and actors in territories that are always specific and
thus rule out management methods that are too mechanical or excessively
standardised. In other words, it seems important to question how an effective
public policy can be produced and how this can be translated at the institutional
level. Today, the challenges associated with decentralisation, the division of
powers between the different levels of government, the holding of multiple
offices (cumul des mandats) and the voting systems cannot be dissociated from
the quest for practical effectiveness when addressing public issues (Bezes and
Le Lidec, 2010). The decentralisation movement that has affected France since
1982 can be characterised as a constituent policy as defined by Lowi (1972), that
is, as an essentially procedural policy that defines the rules of the game and the
323
power of actors. It is typically ‘an institutional reform as a tool of government’

(John, 2011). The biggest problem today is that the division of power does not
cut across the concrete modalities associated with addressing public issues. This
explains why the debate on territorial reform is primarily focused on the division
of powers across government levels.
In addition to the polycentric forms of power linked to both European
integration and decentralisation processes that have developed in all member
states, internal changes that touch on the very nature of public action have
emerged. Public issues correspond to ‘issue areas’ whose frame of reference is
increasingly varied. This explains why there is a management problem arising
from the mismatch between the institutional framework that determines political
authority and a public action that does not necessarily fall within its borders.
Clearly distinguishing between what should fall within the authority of the central
government or, on the contrary, within the power of local authorities has thus
become an immense challenge. Therein lies the problem – which was initially
technical – raised by decentralisation in France. It is worth mentioning that all
federal governments have also attempted, to varying degrees, to accurately define
the division of powers between the state and federal states. However, regardless
of whether the division of responsibilities between federal states and the state is
sectoral or functional, interdependent intergovernmental relations have resulted
in the considerable intertwining of policies because of overlapping responsibilities
and resources such as local and national political and administrative staff. In most
policy areas in the United States, one can observe the existence of a concurrent
jurisdiction between states and federal states that is considerably distant from a
distinct vertical separation of functions; this had led to cooperation between the
different states. In Germany, the famous Politikverflechtung, that is, the interlocking
of policies, also largely blurred the previously well-organised roles. Co-financing
and joint actions had begun to develop by the 1950s.
The diversity of local authorities and the variability of the management areas
of issues raise the question of the architecture of the institutional arrangements
that shape how political power is exercised, as well as the forms of cooperation
between stakeholders. Unquestionably, there has been a disproportionate increase
in coordination problems. Ultimately, by revealing generalised interdependence,
the dispersion of political power has only reinforced the need for cooperation.
France is thus involved in a game of ‘competition and collusion’ in which
no single player holds the keys to the future but, rather, all the players must
learn to deal with increasingly explicit dependent relationships. This presents
a twofold problem: first, the changing nature of public issues greatly increases
interdependency between the different actors. Second, cooperation becomes
difficult and is hampered by the segmentation, fragmentation and multiplication of
the political elites linked to decentralisation. The multiple forms of partnerships,
roundtable discussions and co-financing do no more than illustrate that, although
the playing field is wider, the institutional architecture induces increasingly
prohibitive coordination costs. Moreover, the institutional architecture has also
324
introduced a lack of transparency not only with respect to the processes involved
that will probably remain opaque in the long run, but even more so in relation
to the responsibility of the relevant actors (Duran and Thoenig, 1996; Gilbert
and Thoenig, 1999).
How can the ability to address public issues be understood, given that this
does not automatically depend on positions of authority alone? This question is
particularly true in France, a country which, after long being the most successful
example of legal centralisation, launched a local government reform that led to
the emergence of a political and administrative system without real hierarchy
between levels of government. Indeed, according to French law, one local
authority cannot be under the supervision of another. To the surprise of foreign
observers, France gave birth to a territorial management system where actors are
forced to coordinate without a formally institutionalised hierarchical solution.
This explains why the principle of ‘community leader’ (collectivité chef de file) was
adopted during the constitutional revision of 2003. This solution was reassuring
because it never implies a definitive solution. Indeed, it is simply a management
process that involves deciding when a local authority (municipality, department
or region) should act as a facilitator for the solution to a particular problem by
making arrangements for joint actions. This local authority, however, has no power
to impose these arrangements on other actors. Over and above the explosion
of coordination costs, the multiple scenes and intergovernmental relations
undoubtedly raise the question of prioritisation. Although public issues reflect a
generalised interdependence that is part of ‘multilevel governance’, the diversity
of local authorities and the variability of the management area of public problems
clearly raise the question of the institutional frameworks which are supposed to
determine how political power is exercised and to influence the modalities of
cooperation between the concerned actors. This explains the blurred composition
of authority and the extreme difficulty encountered in the attempt to produce
institutional reforms capable of addressing the coordination of action. A direct
consequence in France has been the current impasse in the local government
reform. The challenge can be articulated as follows: the question that now arises
concerns the development of institutional mechanisms that allow satisfactory relationships
to be built between authority and power. Put differently, one must question whether
the structures of authority hinder the exercise of power, and whether the reality
of power contradicts the statement of authority. How, then, should one reflect
on hierarchy in a context of generalised interdependence?
In addition, because of the European Union, there is a need for governments to
distance themselves from the narrow definition of decentralisation as the simple
confrontation between the state and local authorities due to the fact that the
competition between different territories takes place beyond national borders. For
the first time, questions have been clearly raised about the boundaries of the French
political system and these have introduced a fundamental shift in the country’s
political culture by shattering the image of policy as a magical space where all is
recomposed and the unity of society restored. Political power in France has been
325
haunted by a unitary and community representation of policy embodied in the

greatness of the nation state. Analysis reveals a world lacking a specific centre and
shaped, to the contrary, by multiple centres, heterogeneous ends and a variety
of actors. The French idea of power is now shifting. However, the pervasiveness
of the former model is still visible in the lively debates on sovereignty and the
republic. Naturally, the importance of institutional issues has also been affected
by the weakening of the state model.
It is clear that public policy analysis obliges actors to consider the discrepancies
and, thus, the tensions that may appear between institutionalised political
territories and the spaces where real governance occurs, and between the
widespread support political authorities need in order to act and the particularist
and partial legitimacy necessary for the management of specific public policies.
While there is a need to reflect on the differences between local institutions and
the requirements of a given action which may overwhelm and thus potentially
weaken these institutions, it also seems important to reflect on the issue of
legitimacy that must be addressed by public authorities who attempt to justify
their legitimacy through their actions. These different aspects must be taken into
account and linked together within a renewed theory of democracy.
Rebuilding democracy
Today, this implies reflection on how democracy might be adapted to the reality
of political power which has undergone great transformation and into which
public policy analysis has provided new insights. Political power must incorporate
a new dimension associated with its ability to address community problems.
According to the so-called two-dimensional definition of democracy, a good
government now needs to be responsive, that is, sensitive to social demands, and
problem-solving, that is, capable of effectively addressing community problems
(Scharpf, 1997; Leca, 1997).
Redefining the categories of justification of political power leads to the
emergence of new constraints on the exercise of power. Governors’ demands for
legitimacy are no longer simply tied to the legality of their actions but, rather,
to the impact of these actions as well. In most modern states where efficiency
and performance have become the slogans of the state action doctrine, it
seems necessary to highlight the outcomes of this action. This is indeed a dual
requirement involving research and the correction of mistakes on the one hand,
and democracy on the other. The issue of political participation has emerged
from the shortcomings of a ‘pluralist constitutional democracy’, in the words
of Raymond Aron, which have led to the inability of this democracy to ensure
the full legitimacy of political power and, in parallel, its failure to adequately
monitor the actions of this power (Aron, 1965). The widespread support required
by the authorities in order to take action must be accompanied by a democratic
assessment of how this action is driven. Indeed, the rift between different political
spheres seems to have widened: while front stage politics is highly publicised,
326
back stage politics, where the complex intricacies of policy-making are typically
shielded from the public, has remained in the shadows. Consequently, better
governance involves more than greater efficiency and efficacy in the pursuit
of collectively desired ends. It also involves better deliberative processes about
what the ends, means and outcomes of government action or inaction mean
(Participations Journal, 2011). Democracy can no longer be simply considered
an electoral activity. It must also ensure the participation of all those who
may be affected by policy outcomes and promote the impartial monitoring of
government activities through adapted institutions (Rosanvallon, 2008). This
general framework may explain the development of what has been referred to
in France as new forms of political participation, irrespective of whether they
specifically refer to participatory or deliberative approaches of democracy (Duran
and Truong, 2013). In parallel, public policies are now increasingly linked to
the consultation and participation procedures likely to ensure that both citizens
and those for whom policies are intended have a say on public policy; moreover,
these procedures inform governance (Participations Journal, 2012). However, the
manner in which the participation theme has been embodied in the concept of
pluralism has not been without serious ambiguities. It is by no means certain
that what may ultimately arise from negotiation and participation in management
practices should be equated with the development of democracy. The risk lies
in the temptation to endlessly fragment the channels through which subjects
express themselves and voice their demands. This is because governors easily fall
into the trap of seeking within the micro-regulations of so-called ‘participatory’
democracy, a practical antithesis to uncertain political democracy and a specific
way of managing public issues, at the risk of confusing public interest with the
interests of the public.
It thus follows that the government is responsible not only for managing society
but also for building a genuine public sphere. This means that there is more at
stake than simply improving the practice of traditional representative democracy.
Indeed, there is a need to create wider and deeper consultation channels between
citizens and the government with regard to the ends and means of state action
which, more than ever, can be described as a ‘controversy space’. The obligation
to be accountable is the core of responsibility; this obligation must be considered
in relation to the reality of public action, regardless of how difficult this might
prove (Duran, 2013).
Conclusion: political power under tension

How can collective goals whose formulation is rather sensitive, actors with
highly varying statuses and interests, heterogeneous territories and varying
timescales be articulated? This practical question is also a research question that
allows researchers to better grasp how political power is exercised both in terms
of how action is coordinated and in terms of the accountability and legitimacy
of the authorities involved. This reflection renews the issue of the management
327
of public affairs of our societies and questions the centrality of authorities in

regulating society. Reflecting on public action in a situation experiencing the
fragmentation of sovereignty and the dilution of power has led to renewed interest
in the conditions that may enable the formation of a legitimate political order.
This situation is clearly favourable to increased intellectual interest and to the
development of new concepts. In any event, it is clear that public policy analysis
is inseparable from debate on political power. This may be one of the greatest
strengths of French policy analysis which has developed an approach that takes
into account all major theoretical and practical issues that affect all aspects of
political power.
Unlike in many American studies, ‘polity’ cannot be easily stripped of its state
dimension in France. This may explain the extent to which French studies have
remained strongly committed to a line of questioning that does not dissociate
the ability to produce effective public policies from the context that defines
the conditions of possibility. Moreover, public policy analysis provides a good
illustration of the difficulties political power now faces, and it highlights the
substantive issues arising from its structure. Just as no relevant territory exists in
itself, the fact that no scene, in principle, can impose itself on another is also
symbolic; neither the state, nor Europe, nor even the local authorities can do so.
How can one judge this new war of the gods in a situation in which the state no
longer has the monopoly of the public good? What community should be ‘chosen’
as the framework of reference for the actions undertaken by the authorities when
the nation can no longer rely on evidence? While empirical analysis may find
that no one actor dominates another, should public action be reduced to a vast
market of more or less coordinated transactions between autonomous actors?
Which principles should govern how these transactions are coordinated? What
will become of the symbolic dimensions of political order and the formation
of political identities through which actors construct an image of themselves as
political actors? These are clearly the issues in France and they are consistent with
the policy debate even though the precepts of public policy analysis are difficult
to put into practice, as we previously mentioned. Indeed, public policies shape
political power insofar as addressing these questions relates to ‘organisational
policy’, which seeks to address public issues through organisational procedure,
‘constituent policies’, which may determine the rules of collective action, and
‘democracy policies’, whose objective is to provide legitimacy to governors who
must justify both their position of authority and the effectiveness of their actions.
These remarks reflect the concerns about our capacity to steer the society;
indeed, the only real question is whether we can influence reality and the
conditions under which this influence can occur. Preparing for the future thus
depends on our capacity to make the world intelligible. In other words, it depends
on our ability to understand the future as more than just chaos and chance,
and its ability to promote technical and social solutions capable of ensuring its
functioning. Unfortunately, understanding the world does not automatically lead
to a practical mastery of the world. The reflexive nature necessary for developing
328
solutions inevitably encounters the collective dimension of their implementation.

In the words of Wildavsky (1987, 109-203), public policies are indeed a ‘mix of
cogitation and interaction’!
References
Aron, R. (1962) ‘À propos de la théorie politique’, Revue française de science
politique 12(1), 5–26.
Aron, R. (1965) Démocratie et Totalitarisme, Coll. 10/18, Paris, UGE: Union
Générale d’Édition.
Ashford, D.E. (1978) ‘The structural analysis of policy, or institutions do really
matter’, in D.E. Ashford (ed) Comparing Public Policies, Beverly Hills, CA: Sage
Publications.
Badie, B., Birnbaum, P. (1979) Sociologie de l’État, Paris: Calmann-Lévy.
Bezes, P. (2007) ‘Construire des bureaucraties wébériennes à l’ère du New Public
Management’, Critiques internationals 35, 9–29.
Bezes, P., Le Lidec, P. (2010) ‘L’hybridation du modèle territorial français: RGPP
et réorganisation de l’État territorial’, Revue française d’administration publique
4(136), 919–42.
Bezes, P., Le Lidec, P. (2016a) ‘Politiques de l’organisation: Les nouvelles divisions
du travail étatique’, Revue française de science politique 66(3–4), 407–33.
Bezes, P., Le Lidec, P. (2016b) ‘Politiques de la fusion’, Revue française de science
politique 66(3–4), 507–41.
Bezes, P., Pierru, F. (2012) ‘“État, administration et politiques publiques: Les
dé-liaisons dangereuses”, La France au miroir des sciences sociales nord-
américaines’, Gouvernement et action publique 1(2), 41–87.
Caillosse, J. (2016) L’État du droit administratif, Droit et Société, Paris: Éditions
Lextenso.
Dewey, J. (1927) The Public and its Problems, New York: Holt.
Duran, P. (1990) ‘Le savant et le politique: Pour une approche raisonnée de
l’analyse des politiques publiques’, L’Année sociologique 40, 227–59.
Duran, P. (1999) Penser l’action publique (2nd edn), enlarged with a Postface
‘Pouvoir politique, action publique et sciences sociales’, with a preface by Jacques
Caillosse, Droit et Société,Paris: Librairie Générale de Droit et de Jurisprudence
(LGDJ), 2009.
Duran, P. (2003) ‘Le pouvoir politique pris aux mots: Action publique, politique
publique, service public’, Revue Informations sociales 109, 38–49.
Duran, P. (2004) ‘Genèse de l’analyse des politiques publiques’, in Dictionnaire de
l’analyse des politiques publiques, Paris: Presses de Sciences Po.
Duran, P. (2009) ‘Légitimité, droit et action publique’, L’Année sociologique 59(2),
303–44.
Duran, P. (2010) ‘L’évaluation des politiques publique: Une résistible obligation.
Introduction générale’, Revue française des affaires sociales, ‘L’évaluation des
politiques sanitaires et sociales’, special issue coordinated by Patrice Duran,
1–2, 7–24.
329
Duran, P. (2011) ‘Territorialisation’, in R. Pasquier, S. Guigner, A. Cole (eds)

Dictionnaire des politiques territoriales, pp. 475–82, Paris: Presses de Sciences Po.
Duran, P. (2013) ‘La responsabilité administrative au prisme de l’action publique’,
Revue française d’administration publique 147(3), 589–602.
Duran, P., Thoenig, J.-C. (1996) ‘L’État et la gestion publique territoriale’, Revue
Duran P., Truong F. (2013) ‘La participation politique: nouvelles dimensions,
nouveaux problèmes’, Idées économiques et sociales 17, 4–7.
Elmore, R.F. (1979) ‘Backward Mapping: Implementation Research and Policy
Decisions’, Political Science Quarterly 94(4), 601–16.
Gibert, P. (2010) ‘Contrôle et évaluation, au-delà des querelles sémantiques,
parenté et facteurs de différences’, Revue française des affaires sociales 1–2, 71–88.
Gilbert, G., Thoenig, J.-C. (1999) ‘Les cofinancements publics: Des pratiques
aux rationalités’, Revue d’économie financière 51(1), 45–78.
Granger, G.-G. (1995) ‘Les trois aspects de la rationalité économique’, in L.-A.
Gérard-Varet, J.-C. Passeron (eds) Le modèle et l’enquête: Les usages du principe
de rationalité dans les sciences sociales, pp. 567–80, Paris: Éditions de l’École des
Hautes Études en Sciences Sociales.
Halpern, C., Lascoumes, P., Le Galès, P. (2014) L’instrumentation de l’action publique,
Paris: Les Presses de Sciences Po.
Hjern, B., Porter, D.O. (1981) ‘Implementation structures: A new unit of
administrative analysis’, Organization Studies 2(3), 211–27.
John, P. (2011) Making Policy Work, Abingdon: Routledge.
King, D., Le Galès, P. (2011) ‘Sociologie de l’État en recomposition’, Revue
française de sociologie 52(3), 453–80.
Landau, M. (1991) ‘On multiorganizational systems in public administration’,
Journal of Public Administration Research and Theory 1(1), 5–18.
Colin.
Lasswell, H.D., Lerner, D. (eds) 1951The Policy Sciences : Recent Developments in
Scope and Method, Stanford, CA: Stanford University Press.
Leca, J. (1980) Élaboration d’un bilan interdisciplinaire des travaux concernant l’Etat
français d’aujourd’hui et proposition de programmes de recherche, Paris: Commissariat
Général au Plan, Rapport Cordes.
Leca, J. (1997) ‘Le gouvernement en Europe, un gouvernement européen?’,
Leca, J. (2012) ‘L’État entre politics, policies et polity: Ou peut-on sortir du
triangle des Bermudes?’, Gouvernement et action publique 1(1) 59–82.
Le Galès, P., Scott, A.J. (2008) ‘Une révolution bureaucratique britannique?
Autonomie sans contrôle ou “freer markets, more rules”’, Revue française de
sociologie 49(2), 301–30.
Legendre P. (1992) Trésor historique de l’État en France. L’administration classique,
Paris: Fayard.
330
Lowi, T. (1970) ‘Decision making vs policy making: Toward an antidote for

technocracy’, Public Administration Review 30(May/June), 314–25.
Lowi, T. (1972) ‘Four systems of policy, politics and choice’, Public Administration
Review 32, 298–310.
Mockle, D. (2007) La gouvernance, le droit et l’État, Brussels: Bruylant.
Muller, P. (2000) ‘L’analyse cognitive des politiques publiques: vers une sociologie
politique de l’action publique’, Revue française de science politique 50(2), 189–208.
Papadopoulos, Y. (2013) Democracy in Crisis? Politics, Governance and Policy,
Basingstoke: Palgrave Macmillan.
Revue Participations (2011) ‘Démocratie et participation: un état des savoirs’ 1.
Revue Participations (2012) ‘Participation et action publique’ 2.
Rosanvallon, P. (2008) La légitimité démocratique, Paris: Seuil.
Sabatier P. (1986) ‘Top-down and bottom-up approaches to implementation
research: A critical analysis and suggested synthesis’, Journal of Public Policy 6(1),
21–48.
Scharpf, F. (1997) ‘Européanisation et gouvernement démocratique’, Politiques
et management public 15(1), 11–20.
Thoenig, J.C. (1976) ‘Politiques publiques et décisions interorganisationnelles’,
Revue Suisse de Sociologie 2(2), 55–69.
Thoenig, J.-C. (1998) ‘Politiques publiques et action publique’, Revue internationale
de politique compare 5(2), 295–314.
Tilly, C. (ed) 1975) The Formation of National States in Western Europe, Princeton,
NJ: Princeton University Press.
Widavsky, A. (1987) Speaking Truth to Power. The Art and Craft of Policy Analysis,
New Brunswick, NJ: Transaction Publishers, 109–203.
331
Index
Index
Notes: Page numbers followed by an ‘n’ refer to end-of-chapter notes.
A Centre for Consumption Research and

academic economists 290 Documentation (CREDOC) 193
increased importance 285–286 Centre for the Analysis of Regional
international bureaucratisation of economic Administration (CERAT) 53–55
policy 288, 289, 290 Centre for the Coordination and Orientation
international health economics 286–288 of Economic and Social Development
pro-market economic policy 284–285 Research (CORDES) 195, 197, 302
state-centred economic policy 282, 284 Centre for the Sociology of Organisations
academic institutionalisation 52–58 (CSO) 53, 55, 302, 303
academics 217 Centre for the Study of Revenues and Costs
collaboration with practitioners 51 (CERC) 193, 199
disconnect between academic and Charter of Amiens 262, 274
practitioners’ knowledge 8–11 Chirac, Jacques 148
see also policy process studies civil servants 72, 143, 212, 217
ad hoc commissions 201–203 gender unbalance 144
addictology 254 knowledge production 107–110
administration reviews 33 policy analysis practice 110–113
administrative elite see civil servants training 9, 102–107. see also Grandes Écoles
administrative law 39 see also Grand Corps; practitioners; state elites
Administrative Research Centre (CRA) 41 civil service 101
administrative science 31, 32–38 civil society organisations (CSOs) 243–244
uncompleted institutionalisation of 38–44 origins and developments 244–247
administrative task forces 13, 108 policy analysis
anti-poverty policy 255–256 increased use of 251–256
Attali report 289 lack of mutual interest 247–251
civil-society organisations (CSOs) 19–20
B Coe-rexecode 234
bricolage method 91–92 Coeuré, Benoît 288
budgetary choices, rationalisation of (RCB) cognitive approach to policy studies 87
49–50, 51 Commission for the Assessment and
Bureau of Research and Economic Action Monitoring of Public Policies (CEC)
(BRAEC) 270 148–149
bureaucratisation 288–291 Commission on France 202
bureaucratic system 11–12 commissions 201–203, 214
business organisations 19 committees 157, 162–164, 165
Communauté d’université et établissement
(COMUE) 309–310
C
communes 120
Callon, Michel 165
communicational rationality 66
Canada dry comparisons 87
community leaders 324
Catholic confederation (CFTC) 262, 264,
comparative research design 87–89
265, 269–270
compound research designs 89
333
Confederation Centre for Economic and

Social Studies (CCEES) 267–268 education see Communauté d’université et
congresses of administrative sciences 35–36 établissement (COMUE); Grandes Écoles;
Constitutional bylaw on budget acts 2001 universities
(LOLF) 176–177 education policies 126, 231, 272
constitutional reform 140–142 Egger, Rowland 37
consultancy 18, 234 electoral calendar 238
conditions for using 178–180 electoral competition 211–212
development 175–178 electoral cycles 73
impact 180–185 elite networks see Grand Corps
Cooreman, Georges 35 elites 38, 71–72, 85–86, 112, 212–213
co-production of knowledge 165–166 empirical observation 80–82
corporate actors see consultancy employee unions 238
cost-benefit analysis 283 energy policies 238
Council of Economic Analysis (CAE) 288 environmental organisations 250, 254
Crozier, Michel 53, 81, 302 environmental policy 159–162, 250
cultural policy 251 epistemic rationality 319
ethnography see policy ethnography
European Association of Craft, Small- and
D Medium-Sized Enterprises (UEAPME)
data collection methods 82, 84–86 236–237
decentralisation 59, 213, 229, 243, 306, 309, European integration 68–69
323–324 European monetary unification 288
see also territorial administration European policy 236–237
decision-making tools 48–52 European Union 325–326
see also public participation mechanisms expertise
democracy 326–327 economic 196–197, 280–284, 288–291
départements 120–121 economic interest groups 233–236
Deposits and Consignments Fund (CDC) 49 parliament 142–144
differentiation, local 128–129 within political parties 214–215
Directorate-General for the Modernisation of political parties’ use of external 216–217
the State (DGME) 177 social sciences 196–197
think tanks 218–219
E see also consultancy; state expertise
École Libre des Sciences Politiques (Independent
School of Political Sciences) 34, 40, 298 F
ecologist movement 250 Federation of National Education (FEN) 262,
Economic Analysis Council 16, 198–199 272
Economic and Social Research Institute field studies 55
(IRES) 267, 271 Fifth Republic 12, 13, 50–52
economic expertise 196–197, 280–284, Finance Department, Division for Economic
288–291 and Financial Studies 13–14, 110
economic interest groups 226–229 fiscal policy sees economic policy
changes in applied analysis 237–238 Fondapol 218, 219
changes in policy assessment 236–237 French Democratic Confederation of Labour
co-production of statistics and (CFDT) 262, 265–267, 270
macroeconomic forecasting 229–233
neoliberalism and public-private
interdependence 231–233 G
as policy experts 233–236 gender 144, 167, 178, 179, 221n, 239n
economic policy gender equality 112, 178, 179, 291, 266
international bureaucratisation 288–289 gender policy studies 6, 22n, 63, 167,
pro-market 284–285, 286–288 308–309
state-centred 280–284 General Commission for Planning (CGP) 13,
see also academic economists 108–109, 192–193, 302
economic skill 212 General Commission for Strategy and
Prospective 109
334
Index
see also General Commission for Planning internationalisation, policy process studies
(CGP) 56–58
General Confederation of Labour (CGT) 262 internationalisation bureaucratisation 288–
historical development 263–267 291
policy analysis 267–268 inter-party competition 211–212
and internal dynamics 273 interviews 81, 82, 83, 84–86
General Review of Public Policies (RGPP) intra-party policy analysis 213–216
177
Genieys, William 72, 85–86 J
global-sectorial relationship 65 Jivaro comparisons 88
governance 317–318 Jobert, Bruno 64, 65, 81
governance turn 255 Jones framework 86
Grand Corps 11–12, 90, 107–108, 175, 176 Jospin, Lionel 201, 288
Grandes Écoles 9, 40, 102–107, 213, 279, 280, Jospin government 266
284, 309 Joxe, Alain 197
versus universities 296–299, 307
Grenoblois 64–65
K
knowledge production 107–110
H outside the politico-administrative system
Hassenteufel, Patrick 69, 72, 85–86, 87–88 17–20
health economics 282–284, 286–288, public participation mechanisms 157–159
290–291 democratic nature of 166–167
healthcare 86, 111–112, 182–183, 184–185, in environmental policy 159–162
253 impact assessments 170n
Higher Council for Employment, Revenues, lay knowledge 164–166
and Expenses (CSERC) 199 policy process analysis 168–169
higher education see Communauté d’université providing alternative frameworks 167–168
et établissement (COMUE); Grandes Écoles; in regional rail transport policies 162–164
universities see also expertise; state expertise
horizontal dialectic, territorial politics
123–126
L
labour unions 184
I see also trade unions
ideas 64–67 Lagarde, Christine 289
Ideas Centre, Partie Socialiste (PS) 215–216 Langrod, Georges 42
impact assessments 170n Law on Freedom of Association 1901 245
implementation trick 85 lawsuits 252
Independent School of Political Sciences lay knowledge 164–166
(École Libre des Sciences Politiques) 34, 40, learned societies 34–35
298 legislative institutions 147–148
institutionalisation legislative power 139
policy analysis 48–52 litigations 252
policy process studies 52–58 local authorities 183, 324, 325
policy studies 65, 67–71, 81–82 see also territorial administration
Instituts d’Études Politiques (IEPs) 41–42 local differentiation 128–129
integration costs 320 LOLF (Constitutional bylaw on budget acts
intelligence services 145 2001) 176–177
intercommunalité 120
International Federation of Trade Unions
(IFTU) 264 M
international health economics 286–288 macroeconomic forecasting 232–233
pro-market economic policy 286–288 macroeconomics 281–282
International Institute of Administrative management consultants see consultancy
Sciences 35, 37 managerial training 105–107
international policy transfer 70 managerialism 112
335
market-oriented economic policy see pro- P

market economic policy parliament 138
Marshall Plan 48, 49 contributions to policy analysis 145–150
mathematical models 90–91 limitations 138–144
mediators 64–65 power asymmetries 17–18
medical economics 282–284 Partie Socialiste (PS) 211, 212, 212–213, 214,
Members of Parliament (MPs) 143–144 221n
Mény, Yves 305 Ideas Centre 215–216
methodology 79–80, 316 use of external expertise 217
bricolage method 91–92 participatory democracy 238
data collection methods 82, 84–86 see also public participation
empirical observation 80–82 party elites 212–213
mathematical models 90–91 Perroux, François 282
research design 86–89 planning 54
study timeframe 82–84 policy actors 85–86
Ministry of Finance, service of economic and policy analysis
financial studies (SEEF) 49, 50 centralisation 11–15
Mitterrand, François 265 changes in politics and practices 15–20
monetary policy see economic policy disconnect between academic and
Movement for French Companies (MEDEF) practitioners’ knowledge 8–11
227, 228, 234, 238 foundations 38–44
in the French context 3–8
N at the fringe of universities 302–303
National Assembly (NA) 138–139, 140, 142, historical development 1
143, 144, 148 institutionalisation 48–52
National Centre for Scientific Research see also policy process studies; policy studies
(CNRS) 299–300, 307 policy ethnography 82, 85–86
National Foundation for Political Sciences policy evaluation 147–150, 236–237
(FNSP) 40 policy instruments 318–319
National Institute of Higher Internal Security policy knowledge see knowledge production
Studies (IHESI) 197 policy process studies 4–5
National Institute of Statistics and Economic development 47, 52–58
Studies (INSEE) 13, 109, 194, 200 see also policy studies
National School of Administration (ENA) 9, policy states 314
102, 103–104, 106, 212, 220n policy studies 4, 5–8
National Union of Self-employed academisation of 303–310
Professionals (UNAPL) 228, 234 governance turn 255
neo-institutionalism 68–69 institutionalisation 81–82
neoliberalism 70, 194–198, 211, 231–233 institutions and actors 67–71
new public management (NPM) 322 lack of interest in CSOs and NGOs
non-governmental organisations (NGOs) 247–249
19–20, 243–244 methodology
origins and developments 244–247 bricolage method 91–92
policy analysis data collection methods 82, 84–86
increased use of 251–256 empirical observation 80–82
lack of mutual interest 247–251 mathematical models 90–91
normative approach to policy studies 87 research design 86–89
study timeframe 82–84
and politics 72–73
O role of ideas 64–67
Office comparisons 87–88 state elites 71–72
Office for Scientific and Technological see also policy process studies
Assessment (OPECST) 147–148 policy transfer 70
outcomes 321, 322–323 policy-making 66–67
outputs 321 political economy 68
political legitimacy 212–213
336
Index
political participation see public participation rationalisation of budgetary choices (RCB)

political parties 18, 209–210 49–50, 51, 90, 196, 283
inter-party competition 211–212 rationality 319
intra-party policy analysis 213–216 referentials 64–66
political legitimacy 212–213 reform 127–128
use of external expertise 216–217 regeneration 120
political platforms 209 regional rail transport policies 162–164
political power 320–329 regionalisation 125–126
political sciences 39–40 regions 121
politicians 72–73 research design 86–89
political legitimacy 212–213 reviews of administration 33
poverty 255–256
practical rationality 319 S
practitioners Sarkozy, Nicolas 140, 177, 197, 200, 211,
disconnect between academic and 215, 289
practitioners’ knowledge 8–11 schools of administration 33, 40
policy analysis methods 89–92 science of public administration 36–38, 43
see also civil servants Sciences Po 298, 299, 300, 305–306, 309
pragmatic approach to policy studies 87 semi-structured interviews 81, 82, 83, 84–85
problem-solving orientations 8–9 Senate 142
programmatic elites 72, 86 service of economic and financial studies
pro-market economic policy 284–285, (SEEF) 49, 50
286–288 Small Business Act (SBA) 235
protest model 249–250 social change 51
public action 6, 315–320 social dialogue 237–238
and political power 320–329 social science expertise 196–197
public administration, science of 36–38, 43 social science research 299–301
public choice 128 social sciences 51
public debate 161, 165, 169 Société de statistique de Paris 34–35
public hearings 165 societies, learned 34–35
public inquiries 157, 160–161, 165–166, sociology 8–9, 79, 80–81
168–169, 170n sociology of public action 6–7, 8, 315–320
public participation 20, 326–327 Solidarity, Unity, Democracy (SUD) 262
public participation mechanisms 157–159 state elite networks see Grand Corps
democratic nature of 166–167 state elites 71–72, 85–86
in environmental policy 159–162 state expertise 12–15, 176, 191–192, 212
impact assessments 170n development 192–201
lay knowledge 164–166 economics 280–284
policy process analysis 168–169 restructuring 16
providing alternative frameworks 167–168 statistics 229–233
in regional rail transport policies 162–164 Stiglitz-Sen-Fitoussi report 289
public policy 4, 5 Strategy France 199
public policy analysis see policy analysis sustainable development indicators (SDIs) 112
public policy evaluation 147–150
public service 320–323
T
teachers’ unions 272
Q Terra Nova 218–219
quantification 111–113 territorial administration 119–122, 323
vertical and horizontal transformations
R 122–126
rail transport policies 162–164 territorialisation 126–129
railway 168–169 think tanks 18, 218–219, 234, 287
railway committees 162–164 Thoenig, Jean-Claude 53, 81, 303
Rassemblement Pour la République (RPR) 210, timeframes 82–84
213 trade press 232
trade unions 19, 237–238, 261–263, 270
337
Economic and Social Research Institute

(IRES) 271
historical development 263–267
policy analysis 267–271
and internal dynamics 272–273
training
administration 33, 40–41
civil servants 9, 102–107. see also Grandes
Écoles
training policies 125
transnational bureaucratisation 288–291
transport policies 90–91, 125–126, 162–164,
168
U
Union Institute for Social and Economic
Research (ISERES) 268
Union pour un Mouvement Populaire (UMP) 3,
210, 211, 214, 215
think tanks, 219 218
use of external expertise 217
United States
decentralisation 324
Government Accountability Office (GAO)
147
Office of Technological Assessment (OTA)
148
policy analysis 3, 49–50
policy process studies 56–58
science of public administration 36–38, 43
universities 296–299, 302–303, 307
see also academics
users’ committees, rail transport 162–164
V
VAT rates 236
ventriloquist comparisons 88
vertical dialectic, territorial politics 123, 124
voluntary sector 244–247
W
welfare elites 72, 112
women see gender; gender equality
women’s groups 252, 253
338
Vol 11
“French policy analysis has changed dramatically over the past generation, as has
the study of public policy in France. This book is a must-read for anyone seeking to
understand these changes. The coverage is encyclopedic; the contributors define the INTERNATIONAL LIBRARY OF POLICY ANALYSIS
current state of the art; and the editors have brought it together into a coherent SERIES EDITORS:
overview. It is a major contribution to the literature.” Frank R. Baumgartner, IRIS GEVA-MAY & MICHAEL HOWLETT
IRIS GEVA-MAY & MICHAEL HOWLETT

OF POLICY ANALYSIS
INTERNATIONAL LIBRARY
University of North Carolina
Policy analysis in France lays the foundation for a more systematic understanding of policy
analysis in the country. In doing so, the volume discusses the role of the state and its
restructuring, processes of government and governance, and state-society relationships
and policies as both a process and an outcome. Through 18 chapters contributions focus on
policymakers, their practices, ideas and discourses, how they engage in sustained relationships
with a large variety of market and society actors, and the concrete devices they use in order to
make policy objectives operational. This is a comprehensive study of policy analysis in France
that will be valuable to academics and postgraduate students researching and studying a range
of policy and public management areas.
Features of the ILPA series

• a systematic study of policy analysis systems by government and non-governmental actors
POLICY ANALYSIS IN
Edited by Charlotte Halpern, Patrick Hassenteufel and Philippe Zittoun

POLICY ANALYSIS IN FRANCE
• a history of the country’s policy analysis, empirical case studies and a comparative overview
• a key reference for research and teaching in comparative policy analysis and policy studies
Recent volumes published and forthcoming
France
• Policy analysis in Mexico, edited by Jose-Luis Mendez and Mauricio I. Dussauge-Laguna (2017)
• Policy analysis in Belgium, edited by Marleen Brans and David Aubin (2017)
• Policy analysis in the United States, edited by John A. Hird (2018)
• Policy analysis in Canada, edited by Laurent Dobuzinskis and Michael Howlett (2018)
CHARLOTTE HALPERN is associate research professor in political science at Sciences Po, Centre d’Etudes
Européennes (CEE), CNRS, Paris.
PATRICK HASSENTEUFEL is professor in political science at the University of Versailles and Sciences Po Saint-
Germain-en-Laye, Printemps (CNRS).
PHILIPPE ZITTOUN is research professor at LAET-ENTPE (University of Lyon) and General Secretary of the
International Public Policy Association.
www.policypress.co.uk
@policypress PolicyPress
In association with
Edited by Charlotte Halpern, Patrick Hassenteufel
International Comparative Policy Analysis Forum and Philippe Zittoun
and Journal of Comparative Policy Analysis ISSN 2059-0326
Policy analysis in France [HB] [PRINT].indd 1 02/11/2017 10:56

Policy Analysis in France

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Policy Analysis in France

Uploaded by

Copyright:

Available Formats

INTERNATIONAL LIBRARY OF POLICY ANALYSIS

Edited by Charlotte Halpern, Patrick Hassenteufel

Each volume includes a history of the country’s policy analysis which

Early volumes in the series will cover the following countries:

In association with the ICPA-Forum and Journal of Comparative Policy Analysis.

International Library of Policy Analysis, Vol 11

Notes on contributors vii

Part One: The styles and methods of public policy analysis

two On the path to public policy analysis: 31

Part Two: Policy analysis by governments

six Civil servants and policy analysis in central government 101

Part Three: Committees, public inquiries, and consultants

nine Public inquiries, committees 157

eleven The field of state expertise 191

thirteen Economic interest groups and policy analysis in France 225

Part Five: Academic policy analysis

sixteen Economics and policy analysis: ‘from state to market’? 279

Daniel Benamouzig is a research professor in sociology at the National Centre

Sophie Béroud is an associate professor in political science at the University of

Emilie Biland is an associate professor of political science at the University of

Cécile Blatrix is a professor of political science at AgroParisTech, Paris. Her

Laurie Boussaguet is full professor of political science at the University of

Rafaël Cos is a PhD student in political science at Lille University, member of

Guillaume Courty is a professor of political science at Lille University, Centre

Anne-Cécile Douillet is professor of political science at Lille University, member

Claire Dupuy is associate professor of political science at Sciences Po Grenoble

Patrice Duran is a professor at École Normale Supérieur (ENS) de Paris Saclay

Natacha Gally is an associate professor of political science at Université Paris II-

Julie Gervais is an associate professor in political sociology at Sorbonne University.

Guillaume Gourgues is associate professor in political science (University of

Charlotte Halpern is an associate research professor in political science at Sciences

Fabrice Hamelin is associate professor in political science at Université Paris-

Patrick Hassenteufel is a professor of political science at the University of

Mathieu Hauchecorne is an assistant professor in political science at Paris

Frédéric Lebaron is professor at École normale supérieure Paris-Saclay, where

Patrick Le Galès is CNRS research professor in politics and sociology at Sciences

Jacques de Maillard is professor of political science at the University of Versailles-

Marc Milet is assistant professor in political science at University Paris Panthéon

Emmanuel Négrier is CNRS senior researcher at the South European Centre

Renaud Payre is professor of political science at the Université de Lyon, Sciences

Jean-Marie Pernot is senior researcher and political scientist at Institut de

Frédéric Pierru holds a PhD in political science and is sociologist and a

Gilles Pollet is a professor of political science at the University of Lyon/Sciences

Olivier Rozenberg is an associate professor of political science in Sciences Po, at

Andy Smith is research professor at the Centre Emile Durkheim, University

Yves Surel is professor of political science at the University Paris 2 (Panthéon-

Philippe Zittoun is research professor of political science at the LAET-

the work done by policy analysts working in different institutional settings?

Policy analysis in France: introduction

Policy analysis: what it means and how it is studied in the French

Policy analysis: what it means in the French context

The invention of policy process studies in the French context

The emergence of policy studies is closely related to shortcomings of administrative

The institutionalisation of policy studies as part of political science

of scholarly work highlighted the limits of approaches that overestimated the

in order to make policy objectives operational (Commaille and Jobert, 1999;

development of a number of policy studies subfields,9 which together, contribute

Policy analysis in France: an enduring disconnect between

Looking for the forgotten problem-solving orientation

To begin with, we found that there were no notions or concepts available in

and favouring qualitative sociological methods as argued by Claire Dupuy and

Accounting for weak relationships between academics and practitioners

Unlike the situation observed in other countries, where such a relationship