Innovation and Industrial Policies 9781119681021, 1119681022, 9781119681083, 1119681081, 9781119681106, 1119681103

Table of contents :
Cover......Page 1
Half-Title Page......Page 3
Title Page......Page 5
Copyright Page......Page 6
Contents......Page 7
Acknowledgements......Page 9
Introduction......Page 11
1. Industrial Policy and Competition......Page 21
1.1. The State and industrial policy......Page 27
1.1.1. The State organizing the economy......Page 28
1.1.2. Challenging the economic role of the State......Page 36
1.2. Competition policy as the area of public action......Page 39
1.2.1. Control of the market architecture......Page 41
1.2.2. Monitoring the functioning of markets......Page 58
1.3. Conclusion......Page 70
2. Competition and Innovation Policy......Page 73
2.1. The renewal of the framework of thought......Page 74
2.1.1. A new competitive economics......Page 75
2.1.2. The geography of innovation......Page 80
2.1.3. Innovation and competition policy: the Lisbon strategy......Page 85
2.2. Innovation policy as a “new industrial policy”......Page 89
2.2.1. Innovation and territory: competitiveness clusters......Page 90
2.2.2. A new institutional framework for innovation......Page 96
2.2.3. Support for innovative SMEs......Page 98
2.3. The ambiguities of the “new industrial policy”......Page 101
2.3.1. The logic of agglomeration......Page 102
2.3.2. The logic of development......Page 104
2.3.3. A new mode of governance......Page 105
2.4. The Programme d’investissement d’avenir (PIA)......Page 107
2.4.1. The logic of the Juppé-Rocard report......Page 108
2.4.2. The PIA architecture......Page 111
2.4.3. The link between the PIA and the competitiveness clusters......Page 113
2.5. Conclusion......Page 115
3. Reindustrialization Through Innovation......Page 117
3.1. The affirmation of an industrial ambition......Page 120
3.1.1. A new concern: industry......Page 121
3.1.2. The return of the sector concept......Page 124
3.2. The nature of the industry......Page 127
3.2.1. Problems in defining the industry......Page 129
3.2.2. The question of industrial organization......Page 135
3.3. Towards a renewal of State intervention......Page 142
3.3.1. The new virtues of industrial policy......Page 143
3.3.2. Rediscovering the argument for infant industry......Page 146
3.4. Conclusion......Page 149
Conclusion......Page 153
References......Page 165
Index......Page 181
Other titles from iSTE in Innovation, Entrepreneurship and Management......Page 185

Citation preview

Innovation and Industrial Policies

Innovation between Risk and Reward Set coordinated by Bernard Guilhon and Sandra Montchaud

Volume 5

Innovation and Industrial Policies

Joël-Thomas Ravix Marc Deschamps

First published 2019 in Great Britain and the United States by ISTE Ltd and John Wiley & Sons, Inc.

Apart from any fair dealing for the purposes of research or private study, or criticism or review, as permitted under the Copyright, Designs and Patents Act 1988, this publication may only be reproduced, stored or transmitted, in any form or by any means, with the prior permission in writing of the publishers, or in the case of reprographic reproduction in accordance with the terms and licenses issued by the CLA. Enquiries concerning reproduction outside these terms should be sent to the publishers at the undermentioned address: ISTE Ltd 27-37 St George’s Road London SW19 4EU UK

John Wiley & Sons, Inc. 111 River Street Hoboken, NJ 07030 USA

www.iste.co.uk

www.wiley.com

© ISTE Ltd 2019 The rights of Joël-Thomas Ravix and Marc Deschamps to be identified as the authors of this work have been asserted by them in accordance with the Copyright, Designs and Patents Act 1988. Library of Congress Control Number: 2019946976 British Library Cataloguing-in-Publication Data A CIP record for this book is available from the British Library ISBN 978-1-78630-072-0

Contents

Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

vii

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

ix

Chapter 1. Industrial Policy and Competition . . . . . . . . . . . . . . . .

1

1.1. The State and industrial policy . . . . . . . . . . 1.1.1. The State organizing the economy . . . . . . 1.1.2. Challenging the economic role of the State. 1.2. Competition policy as the area of public action 1.2.1. Control of the market architecture . . . . . . 1.2.2. Monitoring the functioning of markets . . . 1.3. Conclusion . . . . . . . . . . . . . . . . . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

. . . . . . .

7 8 16 19 21 38 50

Chapter 2. Competition and Innovation Policy . . . . . . . . . . . . . . .

53

2.1. The renewal of the framework of thought . . . . . . . . . . . 2.1.1. A new competitive economics . . . . . . . . . . . . . . . 2.1.2. The geography of innovation . . . . . . . . . . . . . . . . 2.1.3. Innovation and competition policy: the Lisbon strategy 2.2. Innovation policy as a “new industrial policy” . . . . . . . . 2.2.1. Innovation and territory: competitiveness clusters . . . 2.2.2. A new institutional framework for innovation . . . . . . 2.2.3. Support for innovative SMEs . . . . . . . . . . . . . . . . 2.3. The ambiguities of the “new industrial policy” . . . . . . . . 2.3.1. The logic of agglomeration . . . . . . . . . . . . . . . . . 2.3.2. The logic of development . . . . . . . . . . . . . . . . . . 2.3.3. A new mode of governance . . . . . . . . . . . . . . . . . 2.4. The Programme d’investissement d’avenir (PIA) . . . . . .

. . . . . . .

. . . . . . . . . . . . .

. . . . . . .

. . . . . . . . . . . . .

. . . . . . .

. . . . . . . . . . . . .

. . . . . . .

. . . . . . . . . . . . .

. . . . . . .

. . . . . . . . . . . . .

. . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

54 55 60 65 69 70 76 78 81 82 84 85 87

vi

Innovation and Industrial Policies

2.4.1. The logic of the Juppé-Rocard report . . . . . . . . . . . . . . 2.4.2. The PIA architecture . . . . . . . . . . . . . . . . . . . . . . . . 2.4.3. The link between the PIA and the competitiveness clusters . 2.5. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . .

88 91 93 95

Chapter 3. Reindustrialization Through Innovation . . . . . . . . . . . .

97

3.1. The affirmation of an industrial ambition . . . . . . . 3.1.1. A new concern: industry . . . . . . . . . . . . . . . 3.1.2. The return of the sector concept . . . . . . . . . . 3.2. The nature of the industry . . . . . . . . . . . . . . . . 3.2.1. Problems in defining the industry . . . . . . . . . 3.2.2. The question of industrial organization . . . . . . 3.3. Towards a renewal of State intervention . . . . . . . . 3.3.1. The new virtues of industrial policy . . . . . . . . 3.3.2. Rediscovering the argument for infant industry . 3.4. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . .

. . . . . . . . . .

. . . .

. . . . . . . . . .

. . . .

. . . . . . . . . .

. . . . . . . . . .

100 101 104 107 109 115 122 123 126 129

Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

133

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

145

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

161

Acknowledgements

During the preparation and writing of this book, we had the opportunity to interact with many colleagues and professionals from the public and private sectors. We would naturally like to thank them for their time. The authors also thank their respective laboratories, GREDEG and CRESE, as well as BETA and OFCE-Sciences Po. We would like to express our warmest thanks more directly to those who have agreed to review the entire book, or parts of it, namely: Sylvain Béal (University of Bourgogne – Franche-Comté and CRESE), Julie Beugnot (University of Bourgogne – Franche-Comté and CRESE), Patrice Bougette (University of Nice Sophia Antipolis and GREDEG), Nicolas Mouchnino (SFIC), Julien Pillot (Inseec U, University of Paris-Saclay and RITM), Olivier Sautel (Deloitte Economic Advisory), and Céline Savard-Chambard (University of Lorraine – EEIGM). For this English version of our book, we are indebted to Jenny Helstroffer (University of Lorraine and BETA) and Alexandra Jordan. Of course, none of them, nor their institutions, can be held responsible for the opinions we express or for any errors or shortcomings that may remain.

Introduction

Over the past 20 years, expanding globalization and the development of information and communication technologies have contributed significantly to the idea that innovation is now the key to business’ competitiveness and the growth process of contemporary economies. However, at the same time, constraints created by the internationalization of value chains have gradually transformed controversy over international company relocations into a more alarming debate on the de-industrialization of the French economy (Fontagné and Lorenzi 2005). In addition to the problem of whether such a phenomenon of “industrial disengagement” is “a reality or a statistical chimera” (Cohen and Buigues 2014, p. 17), this debate was exacerbated by the 2008 crisis, the effects of which, first financial and then economic, largely contributed to discredit the idea that markets are still efficient, to the extent that any public intervention in the economic field could only be useless at best, at worst dangerous. Indeed, many economists consider that the crisis has largely demonstrated that markets are not necessarily efficient and the literature makes the argument that without strong government intervention in the industrial sector, market economies could collapse (Stiglitz et al. 2013). This convergence of views leads not only to justifying the introduction of incentives for innovation, and therefore a genuine innovation policy, but above all to call for a return to industrial policy. It thus confirms ex post a relatively long-held intuition which, about 30 years ago, made some economists say of a previous crisis: Science and technology policies have taken over or are taking over from more traditional industrial policies. Faced with the impasses

x

Innovation and Industrial Policies

of the crisis, solutions are sought through technology and innovation (Bellon and De Bandt 1991, p. 839). Once again, the question arises as to whether innovation policy (a modern version of science and technology policies) has now become an essential component of industrial policy or whether it fully replaces it since, as the Commission nationale d’évaluation des politiques d’innovation (CNEPI) states in its 2016 report, “the very idea of innovation policy is in fact relatively recent and results from borrowing from both science and technology policy and industrial policy” (CNEPI 2016, p. 12). In recent years, the concepts of innovation and industrial policy have changed considerably, probably because of the major changes that have affected the world economy, but especially under the influence of new analyses of the determinants of economic growth. In addition to the questions concerning the links between innovation policy and industrial policy, there are now also questions about the new justifications and new methods of public intervention, induced by these theoretical advances, which raise new questions about the procedures for evaluating these policies. The purpose of this book is to propose some answers to all these questions. More precisely, it is a question of clarifying not only the theoretical foundations of each of these policies, but also the analytical problems posed by their articulation and evolution. However, before addressing these questions, which will be the subject of future chapters, it is useful, as an introduction, to begin by going back to the historical evolution of the meaning of the words “innovation” and “industry”. It is indeed possible to note with Benoît Godin (2017) that, contrary to its current meaning and until the 18th Century, the term innovation had a pejorative and fatal connotation. In particular, this observation becomes clear if we consult Antoine Furetière’s Dictionnaire universel (1690), which defines the verb “innovate” as the act of “changing something already established to replace it with a new one” and an “innovation” as the “change of a custom, of something long-established”. This idea of change is perceived negatively because it is mainly associated with the political domain which, at that time, was entirely dominated by the unrest caused by religious wars. That is why Furetière notes: “In good politics, all innovations are dangerous. Innovations in religion lead to schisms, civil wars”.

Introduction

xi

Such a conception is still present in the Encyclopédie of D’Alembert and Diderot where the innovation article, written by Louis de Jaucourt, refers to the idea of “novelty, or important change that is made in the political government of a State, against the use and rules of its constitution. These kinds of innovations are always deformities in the political order” (Jaucourt 1766b, p. 755). In this almost unanimous condemnation, Sir Francis Bacon is in a way an exception, since he was undoubtedly the first to introduce a shift in the 18th Century to relativize the negative and polemical use of the word “innovation”. Indeed, he devotes one of his Essays or Consels, Civil and Moral (1625) to the question of “Innovations”, in which he states that “all this is true if time stood still; which contrariwise moveth so round that a forward retention of custom is as turbulent a thing as an innovation” (Bacon 1625, p. 433). Bacon also advises that we should, like nature, favor slow and gradual innovations over sudden and radical ones: Men in their innovations would follow the example of time itself, which indeed innovateth greatly, but quietly, and by degrees scarce to be perceived. For otherwise, whatsoever is new is unlooked for; and ever it mends some, and pairs others; and he that is holpen takes it for a fortune, and thanks the time; and he that is hurt, for a wrong, and imputeth it to the author (ibid.). It is undoubtedly also because of Bacon’s influence that the idea of progress emerged in the 18th Century, particularly under the action of Turgot and even more so under Condorcet, and that the word innovation began to take on a more positive meaning. Thus, referring to the criticisms addressed to Turgot by his detractors, Condorcet writes: We have not understood in these reproaches the one of loving innovation, because this reproach can only be made in good faith by men left to the most shameful ignorance. One only has to look around to see that all peoples have an urgent interest in seeing great innovations happen. The taste for new things is, like the spirit of the system, one of those vague accusations that fools and rascals never tire of repeating against men who have spirit or virtues. Why innovate, then? said a general farmer

xii

Innovation and Industrial Policies

naively in 1775; are we not well? (Condorcet 1786, p. 156, note 1). More generally, Condorcet often uses the word “innovation” in the various praises he wrote as Secretary of the Academy of Sciences from 1773 and as Secretary of the French Academy from 1782. Thus, with regard to D’Alembert, he indicates that he had attracted the criticism of “those men in whose eyes the truth seems only a dangerous innovation” (Condorcet 1847d, p. 69). Similarly, he notes that Jussieu had shown the usefulness of a thorough knowledge of botany because it was “a reliable guide in the knowledge of remedies, and it could lead to useful innovations in the art of healing” (Condorcet 1779, p. 245). But it is in his praise of Duhamel du Monceau that we find both the old and the new meaning of the word innovation: The art of farming is like manufacturing: all arts that are exercised only to a level barely more than men need. There is no innovation without advances, without risks (Condorcet 1847b, pp. 616–617). About the difficulties encountered by Jacques Vaucanson, Condorcet writes in his eulogy: “The obstacles of every kind that stand in the way of any useful innovation derive their main strength from the prejudices of those to whom we want to do good” (Condorcet 1847c, p. 654). It is because of the introduction of this modern understanding of the idea of innovation that Condorcet is occasionally presented as the creator of the theory of innovation (Billoret 1989). However, the rehabilitation of the term “innovation” is only really achieved by Jeremy Bentham who, in his Traités des sophismes politiques1 (1816) edited by Etienne Dumont, denounces nostalgia as a sophism and denounces using nostalgia “to fight useful innovations or to defend vicious institutions” (Bentham 1816, p. 2). In particular, it devotes its entire fourth chapter to refuting “The fear of innovation”, showing its absurdity:

1 NOTE: Traités des sophismes politiques (Treaties of Political Sophisms) is a book of extracts taken from many works by Bentham, translated into French. No English language version of this collection exists.

Introduction

xiii

If the reason to condemn a measure is its novelty, that same reason should have condemned everything that exists. To say that something is bad because it is new is to say that all things are bad, at least in their beginning; because everything that is old has been new: everything that is establishment has been innovation. By adopting this so-called argument, you are in contradiction with yourself a thousand times a day (ibid., pp. 43–44). Yet, it should be stressed that it took, for the concept of innovation, more than a century to really establish itself in the thinking and vocabulary of French economists. The term is in fact absent from Charles Coquelin and Gilbert-Urbain Guillaumin’s Dictionnaire de l’économie politique (1852); it is not mentioned either in the second edition of the Nouveau Dictionnaire d’économie politique by Léon Say and Joseph Chailley (1900). It was only about ten years later that this notion was really introduced into economic analysis by Joseph Schumpeter with his Theory of Economic Development, the first German edition of which dates back to 1911. But the economic use of the word innovation only began to spread well after 1934, the date the English translation of Schumpeter’s book was published. As Dominique Guellec (2009) points out, it was not until the 1960s that the first work on innovation began to develop under the leadership of Richard Nelson (1959), Kenneth Arrow (1962) and many others (NBER 1962). The semantics of the word “industry” is marked by a different evolution even though, initially, it is not completely unrelated to the word “innovation”. Indeed, questioning the history of this word, Henri Sée (1925) pointed out that it retained its traditional meaning of “invention” or “knowhow” until the end of the 18th Century and that it would therefore be necessary to wait until the publication in 1819 of Jean-Antoine Chaptal’s book, entitled De l'industrie française, to find the word used in its modern sense. However, this point of view was contested shortly afterwards by Henri Hauser (1925) who defended the idea that the new meaning of the term “industry” could already be found in earlier texts, as is the case with Jean-Marie Roland de La Platière in his “Discours préliminaire” of the volume of the Encyclopédie méthodique. Manufactures, arts et métiers (1790). He then reinforced his argument in the last chapter of his book, Les débuts du capitalisme (1927), based on work by Turgot, Lemercier de La Rivière and Baudeau. However, he still failed to convince Henri Sée (1928, p. 326), who considers some of the examples selected to be questionable.

xiv

Innovation and Industrial Policies

A few years later, Paul Harsin summarized this debate by pointing out that in the 18th Century, the word “industry” could in fact have three different meanings. The first is “the classical meaning of invention, of knowhow” (Harsin 1930, p. 235). For him, this is “the common meaning, if not in the spoken language, then at least in written language” (ibid.). In its second sense, “the word is identified with trade, and opposed to agriculture” (ibid., p. 236). Harsin offers several illustrations citing administrative texts, but also writing by Boisguilbert and Vauban. Finally, the third meaning is the “absolutely modern sense of industrial enterprise” (ibid., p. 237). Although these three meanings of the word “industry” coexisted at that time, Harsin concludes: It seems that it was popular usage that, as early as the 18th Century, qualified the word to designate a special branch of economic production: that of the processing of raw materials (ibid., p. 242). However, the polysemy of the word “industry”, identified by Paul Harsin and later taken up by Michael James (1977), remains incomplete. It neglects two other meanings, common in the 18th Century, that link its origin to the Latin industria. For example, the Abrégé du dictionnaire de Trévoux defines the word industry in the following terms: Dexterity, invention; ability to make something, a purpose or a job succeed. It is sometimes considered simply as art, as work. People who survive by their own eloquence, such as scoundrels, flatterers, scroungers, opinion leaders, etc., are called Knights of Industry in a negative sense (Berthelin 1762, p. 568). The first of these two meanings is that of “work”, but this more precisely designates the activity leading to the realization of something. Indeed, Say and Chailley’s Nouveau dictionnaire d’économie politique (1900, II, p. 66) discards the error of taking the word “industry” as a simple synonym for the word “work”, because these two terms have clearly distinct meanings: “Work refers to the pure and simple exercise of man’s physical forces or intellectual faculties”. However, work alone is not enough for production, capital and natural agents must also be involved; and “this set of combinations can only be correctly referred to by the word industry; the term work would not fit it” (ibid., pp. 66–67). Joseph Schumpeter points out that this is an old-fashioned conception since “the scholastics doctors [...]

Introduction

xv

distinguished the businessman’s industria from the workman’s labor” (1954, pp. 554–555). It is in this sense that the word is still used by Jean-Baptiste Say when he speaks of “this intelligent work that is referred to as industry” (Say 1840, I, p. 24); or again, when he indicates that it is necessary “that industrious man possess products that already exist, without which his industry, however skilled it may be assumed, would remain in inaction” (Say 1841, p. 68). This distinction between industria and labor is important for two reasons. On the one hand, it makes it possible to differentiate the activity of the entrepreneur from the work of the simple employee, as Adam Smith indicates in particular: The profits of stock, it may perhaps be thought are only a different name for the wages of a particular sort of labor, the labor of inspection and direction. They are, however, altogether different, are regulated by quite different principles, and bear no proportion to the quantity, the hardship, or the ingenuity of this supposed labor of inspection and direction. They are regulated altogether by the value of the stock employed, and are greater or smaller in proportion to the extent of this stock (Smith 1976, I, p. 66). On the other hand, in this sense of the word, “industry” is also associated with the idea of skill, ingenuity and even cunning. Also, it is not surprising that Richard Cantillon includes “thieves” in what he calls “entrepreneurs who do their own work without any funding” (Cantillon 1952, pp. 31–32). Such a conception then logically leads us to the second meaning of the word “industry”, which is mentioned in the Abrégé du dictionnaire de Trévoux. The latter refers to a forgotten meaning of the term (Fontaine 1992) since it contains, as was already the case for the word “innovation”, a pejorative meaning related to the idea that know-how, dexterity or skill can be associated with dishonest or unfair actions. The idea that industry would be harmful is partially present in the physiocratic thesis of the sterility of industry and commerce. However, although physiocrats do not really condemn these two activities, they nonetheless condemn the dangers they create because they contribute to the development of luxury at the expense of agriculture. While the condemnation is primarily based on economic logic, that of net proceeds, it nevertheless has moral repercussions. Also, more than industry, it is the behavior of the industrialist that is perceived in a negative way.

xvi

Innovation and Industrial Policies

In this regard, Philippe Fontaine (1992, p. 24) shows that it was undoubtedly the development of industrialism at the beginning of the 19th Century and the celebration of industrial virtues that precipitated the elimination of the pejorative sense of “industry”. To illustrate his point, he quotes Charles Dunoyer who, in his 1825 book, L'industrie et la morale considérées dans leurs rapports avec la liberté, writes: Let us be careful not to confuse those who work with people who intrigue, and industrious men, with the knights of industry. If they need more than one vice in order to prosper, those who succeed cannot do without the moral qualities that embody a good man (Dunoyer 1825, p. 105). The links that unite the semantic evolution of the words “innovation” and “industry” appear even more clearly if we refer to the entry for “industry” in the Encyclopédie. Written by Louis de Jaucourt, this article begins by specifying that: industry, taken in a metaphysical sense, is [...] a faculty of the soul, whose object rolls over the mechanical productions and operations, which are the fruit of invention, and not simply of imitation, skill and routine, as in the ordinary works of artisans (Jaucourt 1766a, p. 694). Jaucourt adds that industry is the daughter of invention, because: the quiet and extensive imagination, easy penetration and quick design, give industry. Those who are very industrious do not always have a sure taste, nor a high level of genius. I say more, ordinary geniuses, geniuses who are not very good at researching, discovering, grasping abstract ideas, can have a lot of industry (ibid.). Jaucourt then discusses the areas of political law and trade in which: this word means two things; either simple hand work, or inventions of the mind in useful machines, in relation to arts and crafts; industry sometimes contains one or the other of these two things, and often brings them both together (ibid.).

Introduction

xvii

By associating labor and industria, work and invention, to make them two sides of the same coin, that of industry, Louis de Jaucourt takes up the cause of economic development and refuses to “object to the utility of industrial inventions” (ibid., p. 695). Despite its brevity, this detour through the history of ideas and economic thought shows that trying to explain the links between industrial policy and innovation policy seems to prevent innovation and industry from being separated, which means that the movement, dynamics or even the evolution of economic phenomena must be at the heart of our approach. This idea will serve as a guiding principle for this book since we will deal with the current foundations of innovation and industrial policy, placing them in their historical context. Above all, we will highlight the decisive influence of the evolution of the economic concepts and theories that serve as their justification. In this perspective, the main objective is to show that, in recent years, it is in fact competition policy that has gradually established itself as a unified framework for defining and constraining both industrial policy and innovation policy, particularly at the European Union level. Thus, to the traditional notion of industrial policy stricto sensu, whose objective was to strengthen the competitiveness of companies and economic activities for reasons of national independence, technological autonomy, bankruptcy of private initiative, decline of traditional activities, or even territorial balance, has in fact been replaced by an industrial policy aimed mainly at correcting market failures linked to information imperfections, the existence of externalities and public goods, or the presence of increasing returns to scale, both in research and innovation and in industrial matters. Within competition policy, innovation policy and industrial policy are presented as two particular and most often independent sub-domains. The reasons for this relative autonomy and its implications will be presented and analyzed in detail in the first two chapters. The third and final chapter will be devoted to an analysis of the coherence problems, both theoretical and practical, arising from the definition and implementation of such a concept, which consists of submitting innovation and industrial policy to the criteria of competition policy. The aim will be to understand how to reconcile and articulate a horizontal logic reflecting the need for a sustained effort in favor of innovation and research, and a vertical logic of greater support for industry, both compatible and consistent with the establishment of a competitive environment, favoring the shift from a topdown logic to a bottom-up logic in order to ensure a real change in the way

xviii

Innovation and Industrial Policies

public action is governed. In other words, the new question that emerges is how the recent orientations and options taken by innovation policy are likely to really promote the development of industry, without contradicting the established principles of competition policy. In this perspective, the three areas we propose to develop in this book are: industrial policy and competition (Chapter 1), competition and innovation policy (Chapter 2) and reindustrialization through innovation (Chapter 3).

1 Industrial Policy and Competition

Economic historians generally agree that about nine thousand years ago, the world underwent a first mutation with the beginning of the first agricultural revolution characterized by the domestication of certain animals, which thus allowed mankind to transition from hunter-gatherers into farmers. From the second half of the 18th Century onwards, three industrial revolutions followed one another. The first, generally dated between 1760 and 1840, marked the era of mechanized production with, in particular, the invention of the weaving machine, the steam engine and the construction of the railways. The Second Industrial Revolution, between the end of the 19th Century and the Second World War, allowed mass production thanks to the control of electricity as well as coal and the creation of assembly lines. Finally, from the 1960s onwards, a third industrial revolution took place with electronics and information technology. It is these developments that make it possible to live better in today’s world than at any other time in history. Today, people are doing better, are richer and live longer, as Angus Deaton (2013) points out. In a striking summary, he notes that “today in sub-Saharan Africa, children are more likely to survive to age 5 than were English children born in 1918” (Deaton 2013, p. 8). The historian David Landes (1988, p. 5) had already marked this progress by indicating the following trait: “The Englishman of 1750 was closer in material things to Caesar’s legionnaires than to his own greatgrandchildren”. As an illustration, the correlation between the changing world population and scientific discoveries can be seen in Robert Fogel’s (1999) graph, which is taken up by the World Bank (2008), as shown in Figure 1.1.

Innovation and Industrial Policies, First Edition. Joël-Thomas Ravix and Marc Deschamps. © ISTE Ltd 2019. Published by ISTE Ltd and John Wiley & Sons, Inc.

2

Innovation and Industrial Policies

Figure 1.1. World population growth and some major events in the history of technology between -9000 BCE and today (source: World Bank 2008, p. 108)

Yet, despite this prodigious change, the world today faces two major, interrelated facts that began in the 1970s in industrialized countries: the increase in inequality within each country (Bourguignon 2012; Deaton 2016) and the deceleration of economic growth. This second situation had not yet arisen 10 years ago thanks to developments in emerging countries such as BRIC (Brazil-Russia-India-China), which were driving global growth, even though it was generally agreed that they were only catching up with more industrialized countries. However, since the Great Recession, the world economy has grown at a rate of about 3 to 3.5% per year, which is below the post-Second World War average. In all industrialized countries, the problem of economic growth refers to the question of industry. Once the “engine” of growth, these countries are now experiencing deindustrialization; that is, simultaneously a decline in industry’s share of value-added creation, a decline in the active population in

Industrial Po olicy and Comp petition

3

industryy and a decliine in the maarket share of exports. Frrance offers a topical examplee in this regaard, with thee share of ind dustry decreaasing from 224.6% in 1970 to 14.1% in 20016, as shown in the grap ph in Figure 1.2.

Figure 1.2.. Evolution of the share of in ndustry (includ ding energy) in value add ded for France e between 197 70 and 2016 (OECD ( data)

This deindustriaalization is also a reflected d in employment figuress, which have deeclined from m 15% to 100% of the working w popuulation over the past 20 years, as well as in the trade balance, wh hich has gonee from a surpplus to a o words, in 30 yearss, French inddustry has loost about chronic deficit. In other 0s, it has lost almost haalf of its two milllion jobs annd, since thee early 1970 industrial populatioon (between 1975 and 2012, 2 accordding to INS SEE, the uding constrruction – fe fell from populatiion employeed by indusstry – exclu 6,175 000 to 3,180,000, a drop of 48.5%). French F deinddustrializatioon is also perceptiible, as Jeann-Luc Gaffarrd (2017) po oints out, inn the figuress for the numberr of robots (1.22 per 100 jobs j in the manufacturing m g sector com mpared to many in 201 15) and decommissioninng rates, almost twice as maany in Germ which reveals r that the productivve apparatus is not beingg modernizedd. France would thus t have beccome the leaast industriallized countryy in Western Europe, as Pierre-André Buiigues (2016) notes. This phenomenoon of rapid deeindustrializaation, the connsequences oof which are ecoonomic, sociial, politicall and culturral, explains why, evenn though industrial production has continnued to increease in volum me and addeed value, the evollution of inddustry seems to be associated in the minds of thee French

4

Innovation and Industrial Policies

with plant closures, relocation, job losses, brownfields and the decline of their country. By consensus, economists agree that this evolution has its origin in the conjunction of three main explanatory factors: – rapid progress in industrial sector productivity; – the outsourcing by industrial companies of certain activities to companies in the service sector (the so-called “servitization” phenomenon); – the loss of industrial activity to other territories due to international competition. According to Lila Demmou (2010), the first two factors contributed 30% and 25% respectively to the destruction of industrial jobs over the period 1980–2007, with foreign competition accounting for nearly 45%. While everyone now agrees on the general observation that France is deindustrialized, as evidenced in particular by the reports of Jean-Louis Beffa or Louis Gallois (Beffa 2004; Gallois 2012), the same cannot be said for the assessment of this development. Two blocks face each other, as Cohen and Buigues (2014) point out. On the one hand, some consider this to be a development similar to that of agriculture. At the beginning of the 19th Century, two thirds of production came from agriculture, whereas it accounted for only 2.16% of value added in 2016. However, no one today believes that it is necessary to launch an agricultural reconquest and return to the place it previously occupied in the French economy. A France composed of fewer but better-equipped farmers corresponds, as Agnès Bénassy-Quéré (2012) points out, to a higher per capita income for farmers and cheaper food. Farmers of the past went to the factory and contributed to the development of a mass manufacturing industry leading to technological innovation and lower prices. From this perspective, deindustrialization would be the product of a “law of economic evolution by which, the richer a country gets, the fewer industrial products and the more services it consumes, the more it abandons routine tasks to keep the sophisticated tasks of creation, design, development, but also marketing and commercialization”, according to the presentation of this argument by Cohen and Buigues (Cohen and Buigues 2014, p. 38).

Industrial Policy and Competition

5

In support of this position, we can find in particular the analysis of Jagdish Bhagwati (2007), who advocates an agnosticism in terms of specialization; the analysis by Patrick Artus and Marie-Paule Virard (2008) for whom the weight of industry will inevitably decrease in developed countries as a result of capital migration to emerging countries that have lower production costs, increased domestic demand and significant productivity thanks to technological catch-up; or Augustin Landier and David Thesmar’s (2013) analysis, according to which the higher a country’s standard of living is, the less weight industry has. On the other hand, other authors consider that deindustrialization is a serious and worrying phenomenon because it could not only undermine the potential for innovation, but also lead to a permanent external deficit caused by the difference left over from a deficit in the balance of goods for which the exports of services cannot compensate. This analysis, to which Cohen and Buigues (2014) subscribe, is based on the fact that industry is at the heart of the innovation and R&D process. They cite in particular the analysis developed by Suzanne Berger (2013), according to which deindustrialization can lead, beyond a certain threshold, to the dismantling of a productive ecosystem. Indeed, when it is no longer possible to travel back and forth between scientific laboratories and factories, incremental innovation is prevented. However, if such a spiral starts, France will be marginalized in terms of technological innovation, which will also have an impact on increasing productivity in other sectors, since industry remains the main source of productivity for both agriculture and services. In view of these elements, Cohen and Buigues (2014, p. 66) clearly oppose the fabless fab strategy (i.e. an industry without a factory) and stress that “there is no autonomous development of high value-added services against a background of complete deindustrialization”. In the same vein, Bénassy-Quéré (2012) points out that deindustrialization raises three problems: – services are on average traded less between countries than industrial or agricultural goods; – productivity growth in services is slower than in industry; – services may not have the same capacity to drive growth as industry or agriculture.

6

Innovation and Industrial Policies

In the face of this disagreement over the seriousness and consequences of deindustrialization for the French economy, questions relating to the need, objectives and modalities of an industrial policy are also clearly the subject of heated debate. However, in many respects, it can be considered that Jean Tirole’s (2016) treatment of these issues provides a synthesis of the current dominant view among economists. Three points deserve to be highlighted at this level. First, Tirole chooses to address the issue of industrial policy in a chapter entitled “Competition and Industrial Policy”, thus revealing both a link between the latter and a hierarchy. Second, after stressing the virtues of competition and competition policy, he asks whether competition is always a good thing, and writes clearly: “The answer to this question is unequivocally ‘no’” (ibid., p. 477). Third, he points out that “anyone who asks about ‘industrial policy’ must first think about the nature of ‘market failure’; otherwise, one wonders why the State intervenes. But simply analyzing a market failure is not enough” (ibid., p. 481). Then he adds: There is also evidence that competitively neutral industrial policies appear to be more favorable to growth than other industrial policies (ibid., p. 481). Then comes the author’s final position on industrial policy, which is set out on the basis of an observation and then provides “guidelines”. The observation, taken from Dani Rodrik, is as follows: “Whether or not we like industrial policy, governments will continue to make it and the subject will not disappear overnight” (ibid., p. 488). In practical terms, Tirole then states that the aim is to “make these initiatives as successful as possible, even if we know that our knowledge in this area will continue to evolve in the future” (ibid., p. 488), and proposes to adopt seven guidelines on this subject: 1) identify the reason for the market failure and respond; 2) use independent and qualified expertise to select projects and recipients of public funds (protected ex ante evaluation of policy intervention); 3) be attentive to supply and not only to demand; 4) adopt a neutral industrial policy in terms of competition; 5) evaluate ex post and disseminate the results of the evaluations;

Industrial Policy and Competition

7

6) strongly involve the private sector in risk-taking; 7) understand the evolution of our economies. Tirole’s position on industrial policy may surprise the reader as it contrasts in some respects with the aspirations of industrialists (such as Didier Lombard (2017)), the conclusions of the various official reports on industry over the past 10 years, the positions of political leaders, and the feelings of the French population. However, it seems to us to correspond perfectly to the dominant analysis among economists and reflect developments in industrial policy and competition from the 1970s to the early 21st Century. Therefore, starting from this conception and all the previous elements, the purpose of this chapter is to demonstrate that if competition policy was initially conceived, until the mid-1970s, as one of the cogs of industrial policy, it gradually came to completely condition the State’s sphere of action by introducing its logic and purpose. As a result, industrial policy today has two characteristic features: on the one hand, it is conceived in a hollow form based on competition policy and, on the other hand, it is fragmented in the sense that it no longer comes under the control of an upstream state source, clearly identified because it is centralized, as evidenced by the fact that there is no longer any ministry of industry in France at present. In order to establish this result, we will first explain how industrial policy has been gradually challenged and has gradually atrophied. Second, we will show that this development has simultaneously been accompanied by an increase in the power of competition policy in France and Europe, to the point of becoming the background for all forms of public intervention, which must now be neutral in terms of competition. 1.1. The State and industrial policy In order to understand the reasons for this gradual substitution of competition policy for industrial policy, it is worth starting by recalling how a policy for industrial development was implemented after the Second World War. In a second step, we will then be able to specify how the evolution of economic analysis on the one hand, and certain failures on the other, have led to questions about the legitimacy and effectiveness of the State’s direct action in industrial matters. This last point of view is developed in particular

8

Innovation and Industrial Policies

by Cohen and Buigues (2014, p. 427), who show that each of the elements of the model that enabled Les Trentes Glorieuses (1946–1975) turned around in the long term, making it necessary to invent a new form of industrial intervention because of the transition from a catching-up economy to a knowledge and innovation economy. 1.1.1. The State organizing the economy Traditionally, it has been agreed to consider that France has offered, for at least three centuries, a unique interventionist approach to industrial policy. This is with regard to both the place and role of the State and its instruments of action, starting with the action and influence of men belonging to the major bodies of the State, and in particular the Mining Corps (Stoffaës 2011). The origin of this French singularity is associated with France’s history and more specifically with two of its outstanding figures. The first is that of Sully, Superintendent of Finance of Henri IV, who sought, with the advice of Barthélemy de Laffemas, to develop national production through a whole series of actions ranging from the improvement of infrastructures (backfilled, paved and passable roads) to the abolition of small financial and judicial offices, including the abolition of tolls and the development of plantations and industry (for example, the Gobelins Manufactory). The second historical figure is that of Colbert, Controller General of Finance of Louis XIV, who wished to place the kingdom’s economy at the service of the king, in particular by introducing customs duties, creating national manufactures and promoting developments in technology (with the creation of the Academy of Sciences in particular) and infrastructure (roads and canals). The latter’s reputation is at the origin of the noun “colbertism” and “sums up the specificity of the French approach to economic policy” (Stoffaës 2011, p. 51). It is therefore in the context of this French historical specificity that questions relating to the need for an industrial policy (section 1.1.1.1), as well as those concerning its definition and implementation methods (section 1.1.1.2), must be considered. 1.1.1.1. The need for an industrial policy Since Richard Musgrave (1959), economists have considered that, beyond the sovereign functions, States intervene through economic policies

Industrial Policy and Competition

9

in order to fulfill three functions: allocation, distribution and stabilization. In this context, industrial policy may appear to be a necessity, first because industry plays a crucial role in a country’s economy (there is a correlation between the growth rate of industry and that of the economy; industrial productivity is higher than that of other sectors and it influences the rest of the economy); second, because “spontaneous evolution, resulting solely from autonomous decisions by the various actors in industrial life, does not seem fully satisfactory” (Maillet 1984, p. 5), in particular in the event of a crisis. According to Pierre Maillet, to convince oneself of the need for public intervention in the productive fabric, it is necessary to start by asking oneself what is expected of a country’s productive apparatus and, in particular, of its industrial component. According to him, we expect: First of all, to produce at the lowest cost the goods and services that are demanded by the nation; this is the raison d’être of a productive system. But also to provide jobs that are as remunerative and as stable as possible [...]. Some other objectives that the productive system can help to achieve should also be mentioned: price stability (or at least a low rate of inflation), justice in income distribution, balance of payments and a certain degree of independence from other countries (ibid., p. 20). However, although the market is a “powerful factor in efficiency” for Maillet, it is not in a position to satisfy all these expectations and must therefore be “supplemented by various public interventions designed to improve its functioning” (ibid., p. 28). In other words, the need for state intervention is based on market failures, that is, on the fact that a number of elements move real markets away from markets as defined under perfect competition theory; and therefore the efficiency and optimality of markets that are supposed to be guaranteed by the two welfare theorems do not materialize. Among these market failures, Pierre Maillet identifies five main ones. First, the competitive mechanism “can only function fully if the market is transparent and not compartmentalized and provides information over time” (ibid., p. 30). However, companies do not spontaneously post their prices or scales, do not necessarily respect them on their own, and tend to split the

10

Innovation and Industrial Policies

market through agreements or product differentiation. In addition, he considers that the market only offers a short-term view and therefore mismanages the problem of the articulation and coherence of long-term perspectives – a crucial issue for industry and R&D. Second, companies neglect the externalities they generate both when they are positive (such as the complementary effects that can exist between activities or products) and when they are negative (such as the deterioration of the regional or social fabric caused by plant closures). As a result, their decisions are generally biased in relation to the general interest. Third, the decision-making behavior of companies may not lead to an efficient functioning of the economy. On the one hand, this is because they adopt excessively short-sighted and cautious behavior; on the other hand, it is because competition can be self-destructive, as evidenced by “a certain ‘exploitation’ of the workforce (poor working conditions, unhealthy or dangerous, underpayment of certain qualifications, etc.)” (ibid., pp. 33–34). Fourth, companies are unable to accelerate or slow down certain rates to “facilitate the readjustment and reclassification” of their employees (ibid., p. 35). Finally, “the functioning of a decentralized market economy can lead to imbalances resulting in underemployment, an unbalanced balance of payments and higher prices” (ibid.), as well as to under-investment in relation to the socially optimal level. In the end, this author’s analysis, which is representative of the arguments supporting the need for an industrial policy, concludes with the following idea: For economically highly developed nations, where an increasing number of individuals take their destiny into their own hands, in accordance with a democratically established and state-controlled rule of the game, the appropriate form is a mixed economy, based on autonomous enterprises and a market economy, supplemented by public interventions, of limited scope and on carefully chosen points: it is the role of industrial policy to make these interventions intelligently (ibid., p. 37).

Industrial Policy and Competition

11

1.1.1.2. The definition and implementation of an industrial policy From the above, Maillet (1984, p. 25) concludes that the role of public authorities consists, within the framework of an industrial policy, of intervening in business’ behavior by modifying both: – the prospects for developing business opportunities and – self-financing resources and conditions of access to credit, but also by controlling or influencing price trends through: – the evolution of wages; – promoting technological change. Understood in this sense, it is easy to understand why Yves Morvan (1991, p. 415) points out that industrial policy is “omnipresent in modern economies, insofar as all government actions have a direct or indirect impact on industrial structures and agents’ behavior”. He therefore proposes to define industrial policy as “a set of guidelines coordinated by public authorities aimed at influencing the conditions for determining agents’ attitudes (and/or influencing the attitudes themselves), in order to achieve objectives considered important” (ibid., p. 425). This extensive and central conception of industrial policies is found in Table 1.1, inspired by Bertrand Bellon and Jacques De Bandt (1991, p. 832). Politically, this representation corresponds to the French conception from 1945 to 1984, as can be seen by taking a broad look at the history of industrial policy during this period. With the post-war reconstruction, an era of catching-up began, led by Charles de Gaulle, thanks in particular to the Commissariat général du Plan (General Commisseriat of the Plan) set up in 1946 at the initiative of Jean Monnet. This economic plan, unlike that of the USSR, was intended to be indicative and incentive: it was “the concerted economy”. The Plan was a forum for reflection aimed at bringing all social partners and public authorities together towards common objectives; it meant “reducing uncertainty”, according to Pierre Massé’s formula. The role of the State was nevertheless very strong since the productive apparatus had been quite destroyed, such that basic activities had to be restored and the public sector was extremely important in industry; this was particularly due to the 1946

12

Innovation and Industrial Policies

nationalizations, direct aid to companies and the Commission centrale des Marchés created in 1959. The government thus acted in two ways: through direct actions on the productive fabric through public enterprises and administered finance, and through indirect actions to guide the business environment through the Plan. Science and technological policies

Macro-economic policies Currency

Industrial policies Research

Balance of payments

Public procurement

Restructuring operations

Technological policies

Education

Employment

Defense

Privatization

R&D

Training

Growth

Sectoral plans Regulation/ Deregulation Aid to SMEs Investment aid

Equipment Spatial planning

Regional industrial development actions

Aid for business creation Standards Competition policy

Regional policies

Export subsidies Aid to FDI (foreign direct investment)

WTO International agreements

Tariff and non-tariff protection Trade policies

Table 1.1. Articulation of the different levels of political intervention

With the decree of August 9, 1953, France set up a Commission technique des ententes (Technical Commission on Agreements) and thus began a policy of competition before the German Competition Act and the 1957 Treaty of Rome. However, far from initiating a policy of competition in the modern sense of the term, the French initiative was motivated by the desire to have an additional lever for industrial policy.

Industrial Policy and Competition

13

Indeed, in order to build vertical industrial sectors around national champions, the political power created a body totally under its control, the aim of which was to fight against horizontal agreements that weigh on the price of equipment as well as public budgets, and are a source of inflation. This horizontal coordination was institutionalized during the Occupation through the organizing committees. Thus, with this body, the political powers did not intend to fight against coordination per se, but against coordination that was exercised against the State’s will to modernize France’s industrial structures. Theoretically, this articulation of industrial policy and competition policy as an instrument of the former finds justification in Maurice Allais’ theory of competitive planning and Jacques Rueff’s theory of the constitutional market. Box 1.1. Competition policy as a lever for industrial policy (source: Didry and Marty 2016)

During Georges Pompidou’s presidency, the spectrum associated with industrial policy was still very broad since it: mobilized public procurement, sectoral development grants (through the FDES-Fonds de développement économique et social) and spatial planning, public or semi-public financing channels (specialized banks such as the French Bank for Foreign Trade or the Crédit national) and nationalized companies (particularly in the energy sector, with EDF, the Compagnie française des pétroles, Elf-ERAP-Société nationale des pétroles d’Aquitaine or CdF-Charbonnages de France), in order to build a rationality in the allocation of budgetary or public resources and in the modernization that will ensure the country’s competitiveness (Bonin 2010, p. 398). However, faced with the initial difficulties encountered by this model (steel or shipbuilding crisis), Georges Pompidou entrusted Jacques ChabanDelmas with the task of modernizing French industry by introducing more competition, that is, finding a new balance between the State and markets: this would be “contractualization”. This evolution can be illustrated by Simon Nora’s 1967 report, which proposed a reform of public companies such as RATP, GDF, EDF or SNCF. The latter were to be granted much

14

Innovation and Industrial Policies

greater management autonomy under program contracts. It should also be noted that from May 1970 an Inter-ministerial Committee on Industrial Policy was set up, the purpose of which, together with the Plan, was to offer expertise to the State in order to define France’s industrial strategy. In addition to supporting SMEs and exports, one of the main objectives of Georges Pompidou’s industrial policy was to create national champions (of any legal status) capable of competing internationally (for example, Creusot-Loire). It was also at this time that European cooperation was to enjoy significant success with the creation of Airbus in 1970 and the advent of a European space policy. This period also corresponded to the apogee of the Plan. Yet, as Cohen and Buigues (2014, p. 313) point out, behind this appearance of a uniform industrial policy: three situations must be radically distinguished: the conditions for State intervention and the status of national champions differ according to whether the State is faced with powerful industrial actors (whose structures it intends to define and predetermine), with lame ducks (politically destabilizing), or with the absence of any industrial actor in a sector decisive for national independence – this is then the field in which major projects are chosen. In all three cases, the mechanisms are formally comparable: sectoral plans, subsidies or bonifications for modernization, investment, export, concentration, etc.; in practice, the effects of these tools are radically different. Subsequently, Valéry Giscard d’Estaing did not really deviate from the idea that France should gradually become a real market economy, but that the State should first radically transform the French productive system, in particular through its industrial policy, so that it can be competitive when it competes with other nations. In other words, France first had to face more competition internally before it could compete with the others. This justifies the eviction of foreign firms in favor of national groups, as evidenced, for example, by the fact that the Ministry of PTT (Post and Telecommunications) promoted Thomson’s entry into public telephony and forced the American firm ITT (International Telephone and Telegraph) and the Swedish company Ericsson to sell their French subsidiaries (Bouvier

Industrial Policy and Competition

15

2008, p. 81). This conception is reinforced by the importance of the oil and monetary shocks which underline the fragility of the French productive fabric and its dependence on the outside world. It was also from this period that the State conceived itself too as a “stretcher-bearer”, to use Élie Cohen’s expression, by ensuring the social management of industrial restructuring and reconversion. The election of François Mitterrand, the first socialist to hold the office of President in the Fifth Republic, initially strengthened industrial policy and the desire to create and consolidate national champions. However, the Socialists believed that previous governments had not sufficiently developed a strategic vision and had only relied on large groups. It was therefore necessary to intensify industrial policy and extend it to all sectors and companies. This takeover of the entire productive fabric was carried out in particular by a wave of nationalizations (in particular in the banking sector, but also by major industrial groups such as Thomson, Saint-Gobain, RhônePoulenc, Usinor, Sacilor, Suez and the CGE), the establishment of “mobilizing programs” to better link industry and research, a stronger focus on industry, increased credits, interventions in favor of SMEs, and the development of regional industrial policies within the framework of State– Region contracts. This perspective was to be challenged from 1983, the year that marks the beginning of what is usually referred to as “the turning point of rigor”, due to external constraints – linked to the economically liberal turn taken by the elections of Margaret Thatcher and Ronald Reagan – and François Mitterrand’s desire to remain in the European Monetary System. From 1984, France, within the common market and under the impetus of the European Commission, really converted to a market economy, as illustrated, in particular, by the transformation of the telecommunications markets (Bouvier 2008) and the end of administered financing. The socialist power was also struggling against the first shock of deindustrialization with, for example, the creation of the Comité interministériel de restructuration industrielle, CIRI (Inter-ministerial Committee for Industrial Restructuring). The years 1983–1984 thus marked a “break in industrial policy: it was increasingly reduced to an environmental policy”, as Norbert Holcblat and Michel Husson (1990, p. 84) noted, quoting Élie Cohen, who concluded that “the State no longer claimed to direct the economy or even protect industry from the market”.

16

Innovation and Industrial Policies

From this rough perspective, it emerges that the post-war years up to the early 1980s appeared to be a period of “tight reins” piloting of the economy by the State. However, at the end of this period, we see the first signs of questioning, which were to lead to the fragmentation of industrial policy and its suffocation by the gradual rise of competition policy. 1.1.2. Challenging the economic role of the State The above elements indicate that industrial policy, like all State interventions, is understood within the economic analysis through the notion of market failure. It is because the market does not necessarily lead to the desired objectives that the State intervenes to correct its failures. However, the evolution of economic analysis and the global environment, as well as some resounding failures of industrial policy, will lead to the conclusion that market failures do not necessarily imply state intervention: it is indeed possible that the cure is worse than the disease. Therefore, changing the normative perspective, it will now be considered that the State must intervene when two cumulative conditions are met: the existence of a market failure, and the demonstration that the State will improve the situation. This new conception will involve deconstructing the economic role of the State (section 1.1.2.1) and challenging industrial policy (section 1.1.2.2). 1.1.2.1. The end of the enchanted conception of the State France has traditionally made the State the essential actor in the functioning and transformation of society. However, from the 1970s onwards, economic discipline was to be marked by a challenge to the model of the omniscient, omnipotent and benevolent State. First of all, the State, like all other economic actors, has limited knowledge. This means that it has only an imperfect knowledge of the structure of the economy. Indeed, even if the State has resources that are often more important than private economic agents – which is no longer necessarily guaranteed in the age of the Internet and GAFAM (Google, Apple, Facebook, Amazon, Microsoft) – the fact remains that it knows the structure of the economy only imperfectly. This lack of knowledge is mainly due to two factors. First, as Friedrich Hayek in particular pointed out, the structure of the economy is not a permanent and unchanging factor. The

Industrial Policy and Competition

17

millions of choices made all the time create a dynamic and complex interaction. In addition, studies on chaotic dynamics, evolutionism, computational phenomena or game theory have shown that the results may be highly sensitive to initial conditions. Second, one of the major differences between the social sciences and the natural sciences is that human beings understand that they are being analyzed and that it may be in their interests to lie or withhold information. In other words, assuming that the State is omniscient is an extremely strong hypothesis, since it is implicitly based on a perfect knowledge of not only the structure of the economy but also the interactions that drive it. Moreover, even if an omniscient state is assumed, there is no guarantee that it will also be omnipotent. Indeed, it is now generally accepted that economic policy analysis can be approached using the tools of game theory. The state is then defined by its strategies and payments, and faces other players who also have their strategies and payments. In other words, we are far from the idea that the State would act on society in a purely directive way, as in a command and control system. With the consideration of strategic interactions between agents, the question of the impact of expectations on economic policies arises. Indeed, as Robert E. Lucas pointed out, any change in economic policy leads to a change in agents’ expectations. Therefore, if the implementation of an economic policy measure does not take these changes into account, the results obtained may differ from the desired results. Finally, it is the state’s benevolence hypothesis that has undoubtedly been the subject of the strongest criticism, initially from both the Virginia school (James Buchanan) and the Chicago school (Georges Stigler). In essence, it is a question of pointing out the fact that the State as an entity does not exist, because it would only itself consist of women and men who are homo oeconomicus, that is, beings who seek to maximize their welfare functions and are sensitive to incentives. Therefore, far from considering that politicians are driven by the collective interest, it is a question, first of all, of seeing in them beings who make career choices and seek re-election. It is no longer the general interest that serves as a compass, but the search for a majority electorate, which contradicts the idea that the State has a long-term perspective while the market has only a short-term view. Moreover, since politicians are sensitive to incentives, pressure groups (lobbies) will do

18

Innovation and Industrial Policies

everything possible to exploit and capture the State apparatus in order to serve their interests. The idea is that thanks to public power and the fact that it can impose certain rules through legal mechanisms, it can be interesting for a pressure group to “buy” politicians to get what they want. This may involve the introduction of subsidies, barriers against their rivals or restrictive measures against consumers. In other words, pressure groups would act to direct public funding towards their interests in return for career support for politicians. This risk seems all the more serious as voters have difficulty in observing public decisions and their effects. In the end, as Avinash Dixit (1996) points out, the state appears to be neither benevolent, nor omniscient, nor omnipotent. Since the early 1980s, we have therefore entered a period that could be described as the end of the “enchanted” conception of the State, which is essentially reflected in the idea that the existence of market failures does not necessarily imply State intervention. This has led to the fragmentation of centralized public power, both horizontally with the multiplication of independent administrative authorities, and vertically with the WTO, the IMF or the regions (BénassyQuéré et al. 2017). 1.1.2.2. Challenging industrial policy Following the oil shocks of 1974 and 1978, Thatcher’s rise to power in 1979 in the United Kingdom and Reagan’s the following year in the United States, these different ideas about the place and role of the State, which had remained mainly in the academic sphere, would have a direct influence on political leaders. Thus, in his 1981 inaugural speech, Reagan stated: “The State is not the solution to our problem; it is the problem”. We will thus witness a real metamorphosis in the conception of the State, which now presents itself as “a more or less coordinated assembly of public policies” (Muller 2004, p. 34). This transformation, described as a “neoliberal shift”, will highlight the failures of industrial policy and in particular call into question the policy of “national champions”. More generally, it can be observed that criticism of the French conception of industrial policy is not only the work of economists. It is found, for example, in the work of a sociologist like Michel Crozier (1963) who mobilizes the theory of organizations to denounce more generally a phenomenon of bureaucratization of French society.

Industrial Policy and Competition

19

The Plan Calcul (Calculation Plan) was a major program launched by Charles de Gaulle to provide France with a national IT industry and thus ensure its autonomy from both other countries and foreign manufacturers. This objective seemed all the more strategic as IT had a central place in the field of civil and military nuclear power as well as in the electronics industry – in particular semiconductors – as the Compagnie des machines Bull, the only major French company in the sector (14,000 employees in France in 1960) had come under the control of General Electric in 1964, and as IT was becoming an instrument of State reform, according to Pierre-Yves Baudot (2011). This Plan resulted in very high public subsidies (several billion euros), the incentive to buy French as part of public procurement, the creation of two research centers (IRIA – now INRIA – and LETI) and an industrial complex formed by the CII (created by merger in 1966), SPERAC and SESCOSEM. The Calculation Plan ended in 1975 with Valéry Giscard d’Estaing’s decision, which only acknowledged the commercial and industrial failure of the CII, the central nucleus of the Calculation Plan, and therefore that of the attempt to create a French IT industry. Today, the industrial policy conducted in the IT sector, organized around the Calculation Plan, appears consensually as an archetype of failure of French industrial policy if it is evaluated on the basis of the results obtained in terms of the subsidies and actions that have been carried out. In a way, only INRIA remains from the Calculation Plan. Box 1.2. The Plan Calcul (1966–1975) (sources: Mounier-Kuhn 1994; 1 FXO-23 archives 1966, on François-Xavier Ortoli’s Plan Calcul)

1.2. Competition policy as the area of public action Even before the fall of the Berlin Wall on November 9, 1989 and the collapse of the Soviet Union on December 26, 1991, European countries, including France, chose to be market economies and, as such, began to reform their understanding of the economic role of the State. The period from the 1980s to the end of the 2000s thus marks the definitive abandonment of direct State intervention in the productive system and the advent of competition as the main source of economic regulation. Article 4 1 Available on the website of the Historical Archives of the European Union, hosted by the European University of Florence.

20

Innovation and Industrial Policies

of the consolidated version of the Treaty establishing the European Community in 1997 stipulates in particular that States shall act in accordance with the “principle of an open market economy with free competition”. As Pierre Muller (2004, p. 35) points out: it is because a new global reference framework has emerged – the market reference framework based on the idea that market mechanisms are more effective than public mechanisms in regulating modern societies – that a new vision of the State has in turn emerged and that the French model is no longer a model. This transition is mainly achieved by a disengagement of the State, offset by the rise, from the 1980s, of a new institutional form which emerged in 1978: the independent administrative authority. The term “competition policy” will gradually become meaningless and misleading. Indeed, from 1986 onwards, France gradually neutralized any strictly political dimension, since political action would be considered at best unfit and at worst corrupt. Competition policy would thus gradually become a competitive market regulation: the authorities in charge of this regulation are no longer subject to the State and have no obligation to coordinate with other policies pursued by the public authorities. Once placed under the authority of the State, competitive regulation became autonomous in the 1980s and took over control of market architecture (section 1.2.1), as well as control of market functioning (section 1.2.2). Competition policy has thus become both the composer and the conductor of microeconomic policies since the latter, starting with industrial policy (European competence since 1992) and innovation policy, can only be conceived and developed in areas left free by the competitive regulation of markets since they must ensure that their anti-competitive effects are kept to a minimum. In this development, the impact of European integration is particularly strong for two reasons. On the one hand, it is because European law had previously been given absolute primacy over national law for all European acts with binding effect and because, in addition, some of them (thanks to their direct effect) can be directly invoked by all before the national courts. On the other hand, it is because the period 1980–2000 corresponds to a period of densification of European substantive law, which applies to

Industrial Policy and Competition

21

individuals and companies, both at the level of primary law (that of the Treaties) and secondary legislation (that of the European institutions). 1.2.1. Control of the market architecture Once an auxiliary, competition policy has since the 1980s become the force that controls the existence and limits of markets as well as the space left for other policies. This metamorphosis was achieved in particular through three elements. The first was the affirmation and deepening of a competitive order provided by the freedoms of movement and the new meanings given to the central concepts of competition law (section 1.2.1.1). The second was the deregulation of network industries (section 1.2.1.2). The third was the control of market structures, and more specifically concentrations (section 1.2.1.3). 1.2.1.1. The competitive order The European construction, at the beginning of the 1980s, was already well established. Indeed, it was based under primary law both on the Treaty of Paris of April 18, 1951, establishing the ECSC (European Coal and Steel Community), and the Treaties of Rome of March 25, 1957, establishing the EEC (European Economic Community), and Euratom or EAEC (European Atomic Energy Community). The main feature of the European project was already there: a functionalist method based on the economy and the will to build institutions that would transcend nation states in order to build a large European market. This method led in particular to the establishment of a customs union on January 1, 1968, unanimous decision-making from the Luxembourg compromise of January 29, 1966, the creation of the European Council at the Paris Summit on April 9 and 10, 1974, and the election of the European Parliament by direct universal suffrage from June 1979. However, it was between the 1980s and the early 2000s that European integration deepened both in terms of economic and institutional integration and in terms of powers, with four treaties: the Single European Act (1986), the Maastricht Treaty (1992), the Amsterdam Treaty (1997) and the Nice Treaty (2001). Among the major changes resulting from this foursome are: the completion of the internal market without frontiers, qualified majority decision-making by the Council of the European Union, the extension and

22

Innovation and Industrial Policies

enlargement of common policies, the creation of the CFI (Court of First Instance of the European Communities), the creation of the European Union, the single currency and the creation of the ECB (European Central Bank). This renewal of European integration was mainly based on the conclusions of the Fontainebleau European Council of June 25 and 26, 1984, those of the report of the ad hoc committee on institutional matters (chaired by James Dooge) and those of the first report of the ad hoc committee on a people’s Europe (chaired by Pietro Adonnino) of March 29 and 30, 1985, but above all in the European Commission’s White Paper of June 14, 1985, adopted at the initiative of its President Jacques Delors. The conclusion of this document is very clear: Europe stands at a crossroads. We either go ahead – with resolution and determination – or we drop back into mediocrity. We can now either resolve to complete the integration of economies of Europe, or through a lack of political will to face the immense problems involved, we can simply allow Europe to develop into no more than a free trade area (European Commission, 1985, p. 8, section 2019). Competition policy therefore presents itself as the central and structuring policy for reaching the European market. In the Treaty establishing the European Community, since 1992 the provisions on competition have been located between, on the one hand, the principles and foundations and, on the other hand, the various policies (economic and monetary, employment, commercial, social, cultural, etc.). Developments in the 1980s and 2000s led to the clarification of freedom of movement and a rethinking of the main concepts of competition law, in order to legally define the constituent elements of the European market. The main architects of this change have been the European Commission, through its initiative function and general supervisory power, and the European Court of Justice (ECJ), in particular through the preliminary ruling mechanism. Placed at the heart of the completion of the internal market, the freedoms of movement of goods, services, persons and capital correspond both to the main principles of European law and to fundamental freedoms. Since the beginning of European integration, they have been the precondition for

Industrial Policy and Competition

23

effective competition and are based on the principles of market access and non-discrimination on grounds of nationality. The ECJ points this out in particular in its Schul judgment of May 5, 1982 (Case C-15/81, section 33): The concept of a common market as defined by the Court in a consistent line of decisions involves the elimination of all obstacles to intra-community trade in order to merge the national markets into a single market bringing about conditions as close as possible to those of a genuine internal market. The only exceptions to these freedoms are those relating to the system of justifications for obstructing freedom of movement on the grounds of imperative requirements (Articles 36, 51 and 52 of the Treaty on the Functioning of the European Union): public morality, public policy, public security, protection of health and life of humans or animals, plant protection, protection of industrial and commercial property, and protection of national treasures with artistic, historical or archaeological value. These economic freedoms have in particular raised issues relating to technical standards and regulations, professional diplomas and qualifications, freedom of establishment, freedom to provide services and cross-border cooperation of companies. As regards the movement of goods and services on the European market, since there has been a customs union since July 1, 1968, the general principle is that all forms of trade barriers should be prohibited, whether by taxes that have an equivalent effect to customs duties or by measures that have an equivalent effect to quantitative restrictions. The classic de minimis legal principle, according to which the law should not concern itself with small things, is inapplicable in this area. Traditionally, European law had chosen the path of harmonization by using the principle of mutual recognition of national standards and by using standardization in certain areas. From the 1985 White Paper, the European Commission adopted a “new approach”: the Commission defines the essential safety requirements (as well as those relating to other collective interests) for placing a product on the market, the standardization bodies propose ways to achieve these objectives, and all States must recognize a presumption of conformity with the essential requirements when products and services comply with these standards. This approach has also introduced a procedure for alerting and

24

Innovation and Industrial Policies

informing about technical regulations envisaged by Member States. The completion of the internal market is thus based on harmonization (limited to essential requirements) and reference to voluntary standards adopted by European standardization bodies. This development is reinforced by the Single European Act of 1986, which replaces unanimity by a majority in European Council decisions on European regulatory and technical harmonization, with a view to eradicating all forms of trade barriers. It is also within this framework that the Community Customs Code was adopted with Regulation No. 2913/92/EEC of October 12, 1992, which was the first codification of standards relating to trade between European countries and third-party countries. The ECJ, the ultimate guarantor of compliance with European law, confirms this strengthening of the free movement of goods and services within the European territory with numerous case laws. At least three can be mentioned. In its Campus Oil judgment of July 10, 1984 (Case C-72/83), the ECJ held that a State cannot protect an economic sector to avoid the effects of the European treaties, while recognizing that it may take a discriminatory and protectionist measure if it is justified and proportionate on grounds of public security. With the Keck and Mithouard judgment in principle of November 24, 1993 (Cases C-267/91 and C-268/91), the Court introduced a distinction between measures relating to the conditions to which products must conform (e.g. shape, size, weight), which are likely to constitute measures that have equivalent effect, and measures relating to the conditions of sale (e.g. place, advertising, authorized operators) which are likely to constitute measures that have equivalent effect only if it is demonstrated that they affect imported products more than domestic products. Finally, enshrining the European Commission’s new approach, the ECJ in its CIA Security judgment of April 30, 1996 (Case C-194/94) states that a national technical provision which has not been notified is not enforceable against individuals if it has the effect of hindering the exploitation of a product or the provision of services. With regard to the movement of persons, whether natural or legal, it is necessary to allow borders to be crossed in the broadest sense. On the one hand, it is a question of ensuring that each person can freely establish themselves or be employed throughout Europe without facing unjustified obstacles. On the other hand, it must be ensured that each economic operator can freely offer its services to customers located in other Member States,

Industrial Policy and Competition

25

while remaining based in a single State. In other words, it is a question of eliminating all barriers to entry and exit within the internal space, for workers and companies alike, in order to mechanically increase the number of competitors. In addition to these contributions relating to the creation of an area of free movement, the European Commission and the ECJ have, during the years 1980–2000, deeply rooted the competitive order by defining the central concepts of competition law and those relating to public service. Indeed, it is through a jurisprudential construction, since the Treaties do not provide any definition, that the central concepts of business, economic activity and serving the general interest have been defined. The European competitive approach to the notion of “enterprise” is a functional and not an organic approach. In other words, the notion of enterprise goes far beyond commercial companies. Thus, an inventor who makes use of a patent license, a religious community, an economic grouping, a trade union, a professional association or any entity governed by public law may be qualified as a business within the meaning of European competition law if it does not act within the framework of its prerogatives as a public authority (for example, a business entrusted with a certification mission which does not exercise any standardization mission and does not have any decision-making power related to the exercise of public authority). Gradually, the ECJ came to a definition which it first set out in its landmark judgment of Höfner and Helser of April 23, 1991 (Case C-41/90, section 21): “every entity engaged in an economic activity, regardless of the legal status of the entity and way in which it is financed” must be regarded as a business. Consequently, all entities that do not carry out an economic activity are excluded from the application of the competition rules. The ECJ recognized, for example, in its judgment of Poucet and Pistre of February 17, 1993 (Case C-159/91 and C-160/91, section 20) that bodies responsible for the management of social security schemes perform a function of an exclusively social nature. Similarly, in its Eurocontrol judgment of January 19, 1994 (Case C-364/92, section 30), it recognized that an international organization which carries out a mission in the general interest on behalf of the Contracting States, for example airspace control and policing, should not be considered as a business.

26

Innovation and Industrial Policies

However, while these two examples may seem obvious, there are cases where this is not the case. Indeed, Article 90 of the EEC Treaty (Article 106 TFEU) sets out situations where the conclusion on the application (or not) of the competition rules is more difficult to reach. These are cases relating to, respectively, public undertakings and undertakings to which special or exclusive rights are granted (Article 90(1)), as well as undertakings entrusted with the operation of services of general economic interest or that have the character of a tax monopoly (Article 90(2)). Without going into the details of this thorny issue, which has been the subject of many debates, comments, decisions and rulings, two essential points should be mentioned. First, it should be pointed out that the European Commission has provided a framework for the transparency of financial relations between States and public companies, through a series of acts begun with Directive 80/723/EC of June 25, 1980. This framework specifies in particular that the definition of what a public company is must be assessed according to an extensive approach. This means covering cases where there is a direct or indirect influence – even potential, but dominant – of the State (in the broadest sense), whether it owns all of it, has a financial participation or determines the rules governing the company in question. Second, it should be pointed out that it is in this context that questions concerning the definition and limits of SGIs (Services of General Interest) and SGEIs (Services of General Economic Interest) are raised, which make it possible to deal with the notion of public services as it is understood in France. Finally, to conclude on the competitive order, this time strictly at the French level, it should be noted that competition policy was still essentially a matter for the State in the early 1980s. The major change was to take place with the Ordinance of December 1, 1986 on free prices and competition. The two main points concerning this ordinance are, on the one hand, the clear affirmation of the general principle of freedom of prices from which freedom of competition derives, and, on the other hand, the creation of the Conseil de la concurrence (Competition Council). Unlike its predecessor, the Commission de la concurrence, the latter has decision-making powers in matters of agreements and abuses of dominant positions, under the supervision of the judicial authority (Cour d’appel de Paris and Cour de cassation). In addition, by respecting the principle of primacy, the Conseil de la concurrence may cumulatively apply European provisions on anticompetitive practices (Articles 85 and 86 of the EEC Treaty) with those of national law and thus impose a double sanction.

Industrial Policy and Competition

27

1.2.1.2. Deregulation of network industries In terms of economic analysis, the 1980s were marked in particular by the questioning of the traditional idea that network industries (electricity, telecommunications, post, railways, air transport, etc.) could not be competitive because they were entirely natural monopolies. It is the theory of contestable markets that makes it possible to dismiss this idea by emphasizing the notion of potential competition, drawing on not only analyses of network externalities, but also the idea previously developed in the 19th Century by Edwin Chadwick, then by Léon Walras in the field of railways, according to which the absence of competition on the market does not imply the absence of competition for the market. On the basis of these elements, various problems are then identified, including the social inefficiency linked to the dimensions of monopolies, the price and quality of services provided to consumers and businesses, and the issues relating to the financing of the public enterprises to which these monopolies are usually entrusted. The spread and importance of these ideas, as well as the growing importance of network industries in our societies, the political rise of neoliberalism, technological change and the desire to complete the internal market, led to the deregulation of network industries in the European area in the 1990s. This is, of course, another major change in the tools traditionally used by States to implement an industrial policy, particularly for France. In order to trace most of these changes, it is necessary to return successively to the role of the European Commission in this area, the liberalization of these industries, the transformations of the incumbents to which they have led and the advent of sectoral regulation, emphasizing the major role played by the Conseil de la concurrence in the French context. With a view to the completion of the internal market, the European Commission undertook, from the early 1990s, to open up network industries within the European territory to competition. However, the possibility of taking such an initiative, which was contested by several States, was recognized by the ECJ on two grounds. First, with the judgment of France versus Commission of the ECJ of March 19, 1991 (Case C-202/88), the European Commission was granted the ability to enact general legislative acts; in other words it was granted regulatory power. Second, the judgment of the Netherlands versus Commission of the ECJ of February 12, 1992 (Cases of the C-48/90 and C-66/90, section 28) states that the European

28

Innovation and Industrial Policies

Commission has a general power of supervision and may therefore “indicate what measures the State to which the decision is addressed must adopt in order to comply with its obligations under Community law”. These included the Directives on telecommunications terminals (1991), electricity (1996), mail (1997) and gas (1998), as well as the 2001 White Paper on transport. The liberalization of network industries is fundamentally based on the idea that they were organized at that time around an overly broad and therefore erroneous understanding of the notion of natural monopoly. It is therefore necessary to separate within them those activities that are genuinely part of a natural monopoly from those that can be carried out in a competitive manner. As François Lévêque (2004) points out, this opening to competition was carried out in two distinct ways. The first, notably followed by the United States and English-speaking countries, consisted of dismantling the monopolies into several independent entities in order to create several companies. The topical example is the dismantling of the American Telephone and Telegraph (AT&T) operator in 1982, following a competition lawsuit that gave rise to more than a hundred companies. It should be noted that this does not necessarily imply privatization in the case of a public company, as the example of the Swedish railways shows. The other possible path, followed by the European Union, is to make competition possible by removing the barriers that prevented the entry of new entrants and their potential exit. In this context, the incumbent operator is not necessarily dismantled vertically – it retains its borders, but is subject to new requirements. Two main transformations will therefore be imposed on each public incumbent operator whose market is liberalized. The first is to open access to its infrastructure to new operators, whether national or European, private or public. The reasoning is as follows. On the one hand, only the infrastructure, that is the network, corresponds to a natural monopoly because it is not economically possible or socially desirable to reproduce it. On the other hand, since, from a historical point of view, it is the taxes, duties or charges that have paid for this network, it is normal that it should be fairly accessible to all operators. It should also be noted that in this context, the European Commission, then the national competition authorities, have found a particular field of application in network matters for a new form of abuse of dominant position resulting from acculturation, from an American jurisprudential construction that began on

Industrial Policy and Competition

29

April 22, 1912 with the Terminal Railroad decision of the Federal Supreme Court: the theory of essential facilities or essential infrastructure theory. This theory essentially argues that when a dominant operator controls an infrastructure that is essential to the expression of effective competition, in the sense that there is no current or potential substitute, the dominant operator must, as far as possible, provide access to it on reasonable terms. The second transformation is deduced from the first because it then becomes essential to separate, at least in accounting terms, the activities that are part of the network from those that are part of the use of the network within the public incumbent operator. The objective here is both to prevent the incumbent operator from using its monopoly activity (on the network) to finance or promote its competitive activity (the use of the network), and at the same time to allow new operators, but also the incumbent operator, to pay the real cost of their use of the network; this is the question of access charges or network interconnection charges, in order to finance its maintenance and renewal. The aim is to obtain transparency of costs so that they can be distributed fairly. In order to ensure the effectiveness of this system, the deregulation of network industries has simultaneously led to the separation of the State from the incumbent operator and the introduction of sectoral regulation. The separation between the State and the incumbent operator is based on the idea that the former should not set objectives or management for the latter that are different from those they should have. In other words, it is a question of preventing the State from relying on or putting pressure on the incumbent operator to help it conduct a microeconomic policy such as industrial policy, or a macroeconomic policy such as growth policy. Therefore, depending on how much confidence one has in the State, two paths have been followed. The first is that of the United Kingdom, which privatized its network industries, thus severing any link between the State and the incumbent operator. The second, the one notably followed by France, was to consider that the State is capable of not using its public network companies politically and of behaving towards them solely as a shareholder concerned about its dividends. On the other hand, it seemed unthinkable that the State should be directly responsible for the day-to-day implementation and enforcement of regulations relating to the liberalization of these markets, as well as for

30

Innovation and Industrial Policies

ensuring fair relationships between the new operators and the incumbent operator. The State was thus not considered to be able to be impartial when its interests are at stake. Several independent administrative authorities were then created in France – some of which have since become independent public authorities – responsible for what is known as “sectoral regulation”. The principle was that each liberalized market should have its own sectoral regulatory authority; this has led, after several changes, to the existence today of ARAFER (Autorité de régulation des activités ferroviaires et routières), ARCEP (Autorité de régulation des communications électroniques et des postes), ARJEL (Autorité de régulation des jeux en ligne), CRE (Commission de régulation de l’énergie) and CSA (Conseil supérieur de l’audiovisuel). In the face of this new environment, three main questions have arisen with regard to these institutions. They concern respectively their independence, their missions and the means of implementing them, as well as their links with the national competition authority. In order to guarantee impartiality, the State was excluded in favor of independent administrative authorities. However, it remained to be ensured that the latter were truly independent of both the State and private economic powers, that is, they were not subject to any conflict of interest. Two main guarantees were then put in place. The first concerns the status of these authorities, which are organically and functionally independent in the sense that they do not come under the organizational chart or the hierarchical authority of the executive, legislative or any other power. No power has the right to give them instructions or even to influence them. The second guarantee concerns the composition and protection of the members of these authorities. Thus, even if they are generally appointed by the executive branch, the members of these authorities are in principle irrevocably appointed for a term of office whose duration is fixed and known in advance. In addition, there is a whole series of professional and political incompatibilities. Defining the tasks and means at the disposal of a sectoral regulatory authority consists of adopting a teleological approach. It is a question of starting from the objectives set by the law that creates it in order to know what the situation essentially is. By way of illustration, when it was created by Act No. 2000-108 of February 10, 2000, the CRE was the Commission de regulation de l’électricité (electricity regulation commission); it was only

Industrial Policy and Competition

31

with Act No. 2003-8 of January 3, 2003 that it became the Commission de regulation de l’énergie (energy regulation commission) as a result of the European liberalization of the natural gas market, achieved with Directive 98/30/EC of June 22, 1998. More generally, it has been highlighted, in particular by Marie-Anne Frison-Roche (2001), that the emergence of these sectoral regulatory authorities is closely linked to that of a new branch of law: regulatory law. The purpose of the latter is to seek a balance between free competition and another imperative principle – such as the monopoly of transport infrastructure, multiannual programming, equality of information, plurality of information, etc. – in order to achieve a balance. A sectoral regulatory authority therefore has the task of building and safeguarding, for a given sector, a balance between opening up to competition and another imperative principle. To this end, it has very broad, unusual and derogatory means at its disposal in relation to traditional public law since, to ensure this balance, it will be responsible for issuing and adapting the regulations governing the sector – a normative power recognized by the Constitutional Council – as well as settling disputes between operators, where necessary by using a power of sanction. The exercise of its powers and their accumulation – contrary to the French tradition of separation of powers – are, in all cases, ultimately controlled by the judicial or administrative courts, essentially with regard to the principles of impartiality, transparency and proportionality. The relationship between competition law and regulatory law raises the question of the relationship between the competition authority and sectoral regulatory authorities. The issue is then that of interregulation, both among sectoral regulatory authorities and between them and the competition authority. Without exhausting the treatment of these relationships, it is simply necessary to bear in mind, as Anne Perrot (2002) points out, that the competition authority – the Conseil de la concurrence in France until 2009 – has horizontal competence (that is, it concerns all sectors), whereas a sectoral regulatory authority has only vertical competence (that, is it only relates to its own sector). In addition, the action of a sectoral regulatory authority is essentially ex ante, unlike that of a competition authority, which is mainly ex post. It was then that the relevant courts came to specify in their judgments how this interregulation should be carried out in practice. In the French context, Frédéric Marty (2004) highlighted, through statistical analysis, the importance of the Conseil de la concurrence in the liberalization of network industries during the 1990s. Indeed, whether for industries that had not yet been liberalized (railways, post offices, etc.), those

32

Innovation and Industrial Policies

in the process of liberalization (electricity) or those already liberalized (telecommunications), the Conseil de la concurrence has played a structuring role with its opinions, protective measures and decisions. This role is partly recognized by this institution since, in its 2000 annual report, it emphasizes that “opinions and decisions given in the context of emergency procedures have often concerned sectors that were formerly monopolistic and that have been the subject of a relatively recent opening to competition”. The general pattern that emerges from this statistical analysis highlights three phases. In the first, the Conseil de la concurrence is asked by the government to provide its opinion on optimal regulation, the objective being to transpose the European directive concerning the specific sector. In the second phase, the Conseil de la concurrence is approached by a sectoral regulatory authority to issue an opinion on the implementation of the regulation. Finally, in a third phase, when the “rules of the game are clear”, the Conseil de la concurrence is understood contentiously, generally by new players on the market, who want to assert their rights against the practices of the incumbent operator. Finally, it appears that the Conseil de la concurrence, like the national competition authorities of the other Member States, as well as the European authorities, have played a fundamental role in the deregulation of network industries, given the very general nature of European directives and the technological and market developments in these industries. 1.2.1.3. Merger control Among the classic industrial policy tools available to States is the establishment of large national companies (private or public): the “national champions”. The underlying idea is that to be a major economic operator on the European and international markets, a state must first of all be large enough. In addition, having a national champion supports the production system by creating a chain of suppliers and subcontractors, thus promoting employment and government revenues. Consequently, the State will seek to promote or, at a minimum, allow a concentration of actors within the production apparatus, so that they benefit from economies of scale and scope, opportunities and new technologies. However, since the 1980s, this idea has clearly faced two obstacles. The first obstacle is the economic analysis, which offers two arguments against this logic. First, market concentration reduces the number of players present

Industrial Policy and Competition

33

and thus risks leading to anti-competitive behavior leading to a deadweight loss; second, the State is not necessarily, as we have already mentioned, a benevolent and omniscient actor capable of “choosing the winners” of tomorrow. The second obstacle is related to the construction of the European market and its performance in an environment of competition and international firms. Indeed, if each European State supports the creation of its national champion, this not only fragments the internal market, but also risks preventing or delaying the natural and desirable creation, by the market, of a real European champion, which is harmful to European consumers and European companies. In this field, it is possible to identify two periods at European level in the treatment of concentrations: before and after September 21, 1990. Before that date, there were two obstacles to the processing of mergers in the internal market. First, national laws on merger control, where they existed, were limited and not homogeneous, as they were limited to the borders of Member States’ respective territories. This could lead to difficulties for both companies and States. Secondly, apart from related areas in the ECSC Treaty (coal and steel), the European Commission had very limited powers since the EEC Treaty did not say anything about merger control. The main reason for this silence was that the Member States were extremely resistant to the idea of giving Europe such control, as well as to the fact that it seemed necessary to restructure the European economy around larger medium-sized companies. This situation therefore led to difficulties at an early stage, which led the European Commission to make proposals, as early as 1965, to establish a merger control regulation in the Council of European Communities (Vandamme and Simons 1990): proposals that the latter systematically refused. This refusal led the European Commission to circumvent this obstacle by using the provisions related to anti-competitive practices to deal with mergers. It was followed in this direction by the ECJ, which handed down two major judgments on this subject. The first was the Continental Can judgment of February 21, 1973 (Case 6/72), by which, while satisfying the company that brought the action for annulment against the European Commission’s decision, it confirmed in terms of principles that Article 86 of the EEC Treaty could be applied in the case of concentrations where the undertaking initiating the operation is in a dominant position. The second was the Philip Morris judgment of November 17, 1987 (Case 142 and 156/84), in which it stated that an acquisition of a shareholding in the capital

34

Innovation and Industrial Policies

of a competing undertaking does not in itself restrict competition, but that this could be the case if its object or effect is to influence the commercial conduct of the businesses concerned, thereby indicating that Article 85 of the EEC Treaty could apply when acquiring a minority shareholding in the capital of a competing undertaking. However, this circumvention did not solve the problem, as many concentrations could not be covered by Articles 85 and 86 of the EEC Treaty because, in particular, they caused the entities to disappear, which posed an insurmountable difficulty in demonstrating coordination; or there was no dominant position, which excluded any form of abuse thereof. Moreover, the European Commission could only intervene after the abusive operation, which made preventive control impossible, and could not grant any exemption in such a case. Similarly, the European Commission could not, in the context of applying the antitrust article, carry out prior checking. The situation was all the more delicate as from the 1973 crisis onwards, mergers developed in Europe (see the various Annual Reports on Competition Policy submitted by the European Commission during that period) with major consequences for employment and market structure. A new period began on September 21, 1990, with the entry into force of Regulation No 4064/89 of December 21, 1989 on the control of concentrations. This is the main change in competition rules since 1957, and the European Commission, in its 1989 Annual Report on Competition Policy, considered it to be “intended to be a cornerstone of competition policy and to make a substantial contribution to the successful completion of the internal market”. To limit ourselves to the essential aspects of this Regulation, it is possible to distinguish its foundations, its scope, its functioning and the means at the disposal of the European Commission to carry it out. Two legal bases are at the origin of this regulation. These are Article 87 of the EEC Treaty on measures implementing Articles 85 and 86, which must be approved by a qualified majority of the members of the Council of the European Communities, and Article 235 of the EEC Treaty, which gives the Council the ability, in the event of unanimity, to give the Community additional powers necessary to achieve its objectives. Thus, far from being imposed by the European institutions, the adoption of a European merger control could only be achieved with the agreement of all the Member States, and in particular France. As a result of this adoption, the Regulation (Article

Industrial Policy and Competition

35

22) terminated the application of Articles 85 and 86 of the EEC Treaty to concentrations with a Community dimension. The scope of the 1989 Regulation is particularly broad both in terms of the types of transactions it covers and in its application. This Regulation covers any takeover, that is, any possibility of exercising decisive influence over the activity of an undertaking (merger, acquisition, or joint venture where it is of a concentrative nature) exceeding certain turnover thresholds (Articles 1 and 3). Thus, any concentration is now either national (if the thresholds are not crossed) or European (if the thresholds are crossed); the European Commission is the only competent authority for the latter, even though in each case, there are derogatory referral possibilities (Articles 9 and 22). However, Member States have the ability to protect their legitimate national interests – in particular those relating to public security, media plurality or prudential rules (Article 21(3)). Finally, in addition to its application to companies (within the meaning of competition law) located in the European territory, transactions carried out outside the European territory by at least one European company, between a European company and a non-European company, or between foreign companies, may also fall within the scope of this Regulation, which therefore has an extraterritorial scope provided that the European internal market is affected. The stated objective of the Regulation (Article 2) is to avoid effective competition being significantly impeded in the European internal market or in a substantial part of it as a result of a change in the structure of businesses. The assessment of the operation is based on the notion of the creation or strengthening of economic power in the sense of a single dominant position (ECJ United Brands judgment of 1978) or a collective dominant position (Article 2(1)). Thus, the prevention of the formation of a dominant position is now being incorporated into European competition law. In its assessment, the Commission will take into account developments in technical and economic progress, if it is to the advantage of consumers and does not constitute an obstacle to competition (Article 2(1)). In terms of its functioning, European supervision takes place in two stages. During the first, the party or parties acquiring control must give notice of the transaction within one week of conclusion of the agreement and in principle wait at least three weeks after the notification (Article 7) before carrying it: this is therefore an a priori control (Article 4). The second stage begins as soon as all the elements necessary for the examination of the

36

Innovation and Industrial Policies

transaction have been received. The European Commission publishes the notification and begins its examination. The latter is strictly limited in terms of time, since it must take a maximum of four weeks2, at the end of which the Commission may conclude either that the operation does not fall within the scope of the Regulation, or that it is compatible with it, or that it raises serious doubts as to its compatibility and merits a detailed examination. The latter, for a maximum period of four months, from the decision following the first phase, then leads to the conclusion either that the transaction is compatible with the settlement, with any amendments made in the form of an undertaking, or that the transaction is incompatible with the settlement, which makes it impossible to carry out the transaction legally. Over the period 1980–2000, the European Commission opposed eight mergers (including the famous ATR/De Havilland case in 1991), but some companies preferred to abandon their projects and others accepted the conditions required by the European Commission. It should be noted that the Regulation establishes an “Advisory Committee on Concentrations between Businesses” (Article 19), in which Member States may submit their comments, and which must be consulted and issue an opinion. To achieve this control, the European Commission has investigative powers, similar to those it has in relation to anti-competitive practices, ranging from simple requests for information to coercion (Articles 11 to 13). It also has powers of sanction since it can impose fines (when deliberately or through negligence there are: omission of notification, violation of the suspension period, inaccurate or incomplete information, information beyond the time limit), penalty payments (in order to compel the provision of information, submit to a verification ordered by decision, execute a charge, apply the measures ordered by the Commission’s decision), separation (if the operation has already been carried out, the Commission orders devolution), or revocation of a previous decision (incorrect information, compatibility obtained fraudulently, non-compliance with the obligations attached to the decision)3.

2 This period may be extended to six weeks at the request of a Member State justifying the dangers that the concentration represents for its market. 3 In order to clarify the main procedural details of Regulation No 4064/89 before its entry into force, Regulation 2367/90 of January 25, 1990 was adopted at the same time. In particular, it

Industrial Policy and Competition

37

This system continued until the in-depth overhaul of the merger control system following the adoption on January 29, 2004 of Regulation (EC) No. 139/2004, which took into account five amendments: – an extension of the Commission’s exclusive competence by increasing the number of transactions to be notified; – a change in the way thresholds are calculated for financial and credit institutions, as well as for insurance companies; – the explicit possibility for companies to make commitments from the first phase of the review; – the strengthening of the distinction between concentrative joint ventures and cooperative joint ventures; – the impossibility of implementing a concentration before obtaining a decision declaring it compatible. At the French level, following on from the 1945 Ordinance, merger control began with Act No. 77-306 of July 19, 1977 on the control of economic concentration and the repression of unlawful agreements and abuses of dominant positions; it was the first legislative text to deal with this subject explicitly. These provisions will be taken up again when modern competition law emerges in France, since they appear in Title V of the 1986 Ordinance (Articles 38 to 44). Three main features characterized French merger control during the period 1980–2000. First of all, it was an exclusively governmental competency: only the Minister in charge of the economy and the Minister responsible for the economic sector concerned could make decisions in this area. This decision was binding and can only be appealed to the Council of State, in particular with regard to respect for the principle of freedom of trade and industry. Second, it was an optional control both for the Minister in charge of the economy (Article 38) and for the companies concerned (Article 40). Finally, control can be a priori (at the request of companies) or a posteriori (at the request of the Minister in charge of the economy). Thus, in matters of concentration, the Conseil de la concurrence has no power and only intervenes, where necessary, at the request of the Minister in charge of clarified how the most relevant product and geographic market should be delimited, how the markets affected by the proposed transaction should be determined, and how Form CO and supporting documents allowing companies to give notice of a transaction should be introduced.

38

Innovation and Industrial Policies

the economy to provide an opinion (the control procedure at the initiative of the Conseil de la concurrence provided for in Article 43 of the Order has never been used). All in all, it appears that if France can still pursue a microeconomic policy, such as industrial policy, during this period, by allowing the productive apparatus to restructure itself or by promoting such restructuring, it is only on the condition that this policy does not affect effective competition on the internal market within the meaning of European law. During this period, this type of policy is therefore not simply abolished but considerably reduced in its scope of application. Its almost complete abolition was achieved with the Law on the Modernization of the Economy of August 4, 2008, which transfers all merger control to the Competition Authority and only grants the Minister for the Economy a residual power of evocation for reasons other than those relating to competition. From then on, mergers have been governed essentially by competition law. 1.2.2. Monitoring the functioning of markets Defining the playground and its rules is not enough to ensure that the referee and players will respect them. This explains why competition policy, this time acting as a market police, ensures that the State and all economic operators, both public and private, do not distort competition defined by the competitive order. However, while all are affected by the prohibition of anti-competitive practices (agreement and abuse of dominant position), States must also comply with certain rules relating to the subsidies they may decide to grant (section 1.2.2.1) and to public procurement law (section 1.2.2.2). 1.2.2.1. The framework for State subsidies It should be recalled that economic theory explains that, in perfectly competitive markets, any government subsidy is inefficient from a social welfare perspective. However, if the Pareto optimum cannot be achieved due, for example, to market failures (imperfect information, barriers to entry, market power, imperfect mobility of production factors, agglomeration effects, etc.), it becomes necessary to seek a second-best optimum (best solution after the Pareto optimum). However, in general, when one of the conditions for achieving the Pareto optimum is not met, it is not optimal for the other conditions to correspond

Industrial Policy and Competition

39

to those that would have been achieved with the Pareto optimum. In other words, in a second-best optimum situation, subsidies can correct inefficiencies related to the imperfect nature of markets, as it is generally inefficient to want to remove all existing distortions. This is one of the foundations of the general principle that a State is never prohibited from providing subsidies to economic operators. Nevertheless, over the period 1980–2000, this general principle was the subject of a framework both at the level of international rules and at the level of European rules. At the international level, it is essentially the measures concerning subsidies under the GATT (General Agreement on Tariffs and Trade) and then the WTO (World Trade Organization) that are binding on European countries. In this area, three main measures should be highlighted. First, the agreement of April 12, 1979 on the interpretation and application of Articles VI, XVI and XXIII of the GATT. This agreement, in practice sometimes referred to as the “Subsidies and Countervailing Powers Code”, does not provide a definition of the notion of subsidy but introduces control mechanisms. Indeed, in this context, it is incumbent on each signatory State to demonstrate, before imposing a countervailing duty, that its domestic industry is suffering or threatened with serious injury and that this injury is caused by subsidized imports or dumped prices by another State. The damage must be demonstrated on the basis of a series of objective criteria. Export subsidies are thus regulated. With regard to purely domestic subsidies, the agreement stipulates that signatories must seek to prevent them from prejudicing the interests of other signatories, or from cancelling or compromising benefits resulting from the General Agreement for another signatory. A notification system is established, as well as a dispute settlement procedure. It is also provided, in the context of a dispute, that States may make commitments, that is, bilateral agreements, to voluntarily restrict exports. It was in this context, for example, that an agreement between the EEC and the United States on steel was signed on October 21, 1982, under which European producers voluntarily restricted their exports to the United States. Second, the agreement of April 15, 1994 establishing the WTO. This agreement first ended the bilateral voluntary export restriction agreements and introduced a limit on the duration of the safeguard clauses (four years in principle and a maximum of eight years). In addition, the agreement specifies that when a country restricts imports to safeguard its domestic

40

Innovation and Industrial Policies

producers, it must in principle give something in exchange to the person (or persons) to whom it is thus causing injury. However, exports from developing countries enjoy some protection against safeguard measures. The agreement also provides that the exporting country (or countries) may request compensation through consultation. If the parties cannot agree, the exporting country may take an equivalent retaliatory measure (e.g. an increase in customs duties). The WTO Committee on Safeguards monitors the functioning of the Agreement and the fulfillment of commitments made by Members. Finally, the WTO Agreement on Subsidies and Countervailing Measures of 15 April 1994, which entered into force on January 1, 1996. This agreement covers both the granting of subsidies and the use of compensatory measures to neutralize the damage caused. It begins by defining the term subsidy based on three elements: – it is a financial contribution (in any form: donations, loans, equity participation, guarantees, provision of goods or services, tax measures, etc.); – it comes directly or indirectly from the public authorities; – it confers a benefit on its recipient(s). The agreement then distinguishes between prohibited subsidies (those relating to exports or the presence of domestic elements) and subsidies that may be challenged before the WTO or be subject to countervailing measures (these are subsidies that are specifically targeted at one or more companies, industries or regions of the national territory). It is also provided that a State party to this agreement may not impose a countervailing measure unless it can prove that there are simultaneously 1) subsidized imports, 2) injury to a domestic industry and 3) a causal link between the subsidized imports and the injury. Finally, Article 25 of the agreement requires signatory States to give notice of all specific subsidies (at all levels of government and for all product sectors, including agriculture) to the Comité des subventions et des mesures compensatoires (Committee on Subsidies and Countervailing Measures). They must submit new complete notifications every three years and, in the meantime, update notifications. These notifications are subject to in-depth examination and discussion in the Committee. As far as European provisions are concerned, subsidies are organized within a system consisting essentially of two components: economic and

Industrial Policy and Competition

41

social cohesion and State aid control. Since the Single European Act of 1986, economic and social cohesion has been a European competency. It was in this context that Regulation No. 2052/88 of June 24, 1988 was adopted. It sets five priority objectives for this grant system: 1) promoting the development and structural adjustment of regions whose development is lagging behind; 2) aid for the conversion of regions seriously affected by industrial decline; 3) fighting against long-term unemployment; 4) the professional integration of young people; 5) speeding up the adjustment of agricultural structures and promoting the development of rural areas. European regions that meet at least one of these five objectives are eligible for support under this policy. The Maastricht Treaty in 1992 changed this policy in three ways: by establishing the Cohesion Fund, by creating the Committee of the Regions (a new European institution to represent local authorities) and by introducing the principle of subsidiarity into this policy. Finally, it is worth mentioning the creation of a sixth objective (aid for sparsely populated regions) in 1995 and, with the Treaty of Amsterdam in 1997, the establishment of closer coordination of national employment policies with the inclusion of a principle of a European strategy in this field. However, it should be pointed out that since Europe as a political entity began to be formed in 1957, there has been a framework for national State aid within the provisions on competition. Part of the legal doctrine considers that this State aid scheme should in fact be seen as one of the elements of European industrial policy. In other words, before having an explicit industrial policy from the 1990s onwards, the European institutions would have implicitly pursued an industrial policy through the competitive framework of national industrial policies. Before discussing the main provisions concerning European State aid control, which is one of the specific features of European competition policy, it is necessary to specify the principles and then the procedures.

42

Innovation and Industrial Policies

The general principle is the control of subsidies to avoid distortions of competition within the European market. Thus, it is not an absolute prohibition of any form of subsidy, but a strict framework designed to preserve the equality of economic operators on the markets, a necessary condition for the realization and functioning of the European project. There is a risk of partitioning the European market if operators consider that an increase in their market share in a Member State will result in the latter’s aid to its national companies being used in retaliation. On the other hand, European integration is not based on competition from States, which would be a waste of public funds. The State aid regime is therefore centered on an incompatibility in principle. It follows from these elements that aid will in principle be declared incompatible with European law if three conditions are cumulatively met. First of all, it must be State aid. While this first condition may seem trivial, it has nevertheless been the subject of the most debate and litigation. Thus, the acceptance has gradually emerged that aid corresponds to a selective advantage granted to an undertaking – private or public – or a sector, without remuneration or with a remuneration that does not correspond to the amount at which the advantage is assessed. This aid must come, directly or indirectly, from State resources. For example, aid may be a grant, an interest-free loan, a soft loan, a financial guarantee, a tax exemption, or a contribution of goods or services on preferential terms. The European institutions have thus developed an extensive conception of the concept of aid, which must be assessed on the basis of its effects and not on its legal form or purpose. This is the origin of the jurisprudential construction of the criterion of “the private investor acting in a free market context” or “the investor guided by market criteria” (the Netherlands versus Commission judgment of March 13, 1985, Cases 296 and 318/82, section 17), which makes it possible to distinguish two situations in the event that a State has shares in the share capital of an undertaking. Either the State behaves like a private investor, and there is no aid, or the State does not behave like a private investor, and in this case there is aid. The ECJ subsequently clarified that this investor should not, however, be considered as a short-term investor interested in immediate profitability (Belgium versus Commission judgment of July 10, 1986, Case C-234/84, section 15, and Alfa Romeo judgment of March 21, 1991, Case C-305/89, sections 20 and 22). On the other hand, not all general, uniform and automatic economic support measures likely to concern all sectors of activity and which do not favor any specific operator

Industrial Policy and Competition

43

are qualified as State aid. Similarly, if a selective advantage corresponds to the counterpart of a public service, it is not aid. Second, the State aid must distort or threaten to distort competition. This condition can be understood in the same way as the provisions on anticompetitive practices, but it is not limited to this meaning. Finally, this State aid which threatens to distort or distorts competition must affect trade between Member States. This condition is taken in a broad sense. Indeed, in its judgment of July 13, 1988 of France versus Commission (Case C-102/87, section 19), the ECJ held that export aid may distort competition because it may lead to domestic production being maintained or increased; this reduces the opportunities for companies from other Member States to engage in export. In addition, export aid to third countries may also affect European trade, given the interdependence of markets, as the ECJ stated in its judgment of Belgium versus Commission of March 21, 1990 (Case C-142/87, section 35). Ultimately, Article 92(1) of the EEC Treaty (Article 107(1) TFEU) concerns the effect on trade between Member States by means of aid from State resources which distorts or threatens to distort competition by favoring certain operators or certain products. However, the same article, in the following two paragraphs, stipulates the existence of two regimes of derogations from the general principle laid down in the first paragraph. Article 92(2) of the EEC Treaty (Article 107(2) TFEU) distinguishes a block exemption scheme for: – aid of a social nature granted individually to consumers (e.g. tax exemption for the purchase of an electric car, regardless of the brand); – aid to repair damage caused by natural disasters or other exceptional occurrences; – aid to certain German regions as a result of the division, which became obsolete from October 1990. Article 92(3) of the EEC Treaty (Article 107(3) TFEU) implicitly establishes that the European Commission has discretionary power to accept other aid – power explicitly recognized by the ECJ since its Phillip Morris judgment of September 17, 1980 (Case C-730/79, section 24). However, on the one hand, the European Commission must give reasons for its decisions

44

Innovation and Industrial Policies

in this matter in order to preserve the rights of the defense and to allow formal review by the ECJ (i.e. verify that the Commission has not exceeded its discretion and has not committed a manifest error). On the other hand, the paragraph of the Treaty indicates the categories in which this discretionary power is exercised: – in the economic development of regions where the standard of living is abnormally low (the Commission specified in a 1988 communication on regional aid that this corresponds to a per capita GDP below 75% of the Community average) or where there is serious underemployment (in terms of structural unemployment); – in order to promote the execution of an important project of European interest (such as national aid for R&D) or to remedy a serious disturbance in the economy of a Member State; – for the development of certain activities or regions if this does not affect trading conditions to an extent contrary to the common interest; – since the Maastricht Treaty, to promote culture and heritage conservation; – within the framework of new categories created by the Council on a proposal from the European Commission. Turning now to the procedural aspects, a distinction must be made between existing aid and projects relating to new aid. In the first case, whether it is aid granted before the EEC Treaty or aid that has been previously accepted, the European Commission exercises permanent ex post control on the basis of the annual reports provided by the States concerned. This control may lead the Commission to modify or withdraw aid. In the second case, it is an a priori control. Any aid project or modification of existing aid is subject to a prior authorization procedure before it can be implemented. The Commission has two months to examine the proposed aid from the moment it has all the information it needs to carry out its examination. At the end of this period, either explicitly or implicitly, the Commission may raise no objection to this aid. Alternatively, it may initiate the contradictory procedure provided for in Article 93(3) of the EEC Treaty (Article 108(3) TFEU), which provides for the collection of observations by a formal notice from the State and the observations of interested third parties, before a decision is generally made within 18 months. This final decision may be positive, negative, conditional or partly positive and partly

Industrial Policy and Competition

45

negative. In the absence of notification, the Commission opens an ex officio procedure, or following a possible complaint. It may then take interim measures to prevent or suspend the granting of the aid and order the recovery of the sums paid to the company by the State. Any aid that is incompatible with European rules may thus be recovered by decision of the European Commission. This recovery procedure is governed by national law and is accompanied by the payment of interest calculated on the basis of the commercial rate from the date on which the aid was granted. This recovery is not considered as a sanction, but as a corrective measure in the face of found illegality. The obligations to notify and suspend the payment of State aid have a direct effect and, as such, must be respected by national courts ex officio or at the request of an interested third party. However, they cannot decide on the compatibility of the aid, as this falls within the exclusive competence of the European Commission. National courts must therefore block the aid by interim measures, pending the Commission’s decision, or ensure that the Commission’s decision is implemented. In addition, as a result of the direct effect, the State incurs liability action. During the years 1980–2000, the Commission systematically refused any aid relating to an industrial management policy, as well as aid concerning the operation of a company, or a fortiori several if it is determined selectively. It only accepted aid to mitigate the social consequences of company closures. According to the Commission, the industrial strategy must be defined within the market. Aid to a sector in crisis can only be granted in support of restructuring measures to restore competitiveness and only if this is beneficial for the Community and is not at the expense of competitors. According to the European Commission, sectoral policies only delay the necessary structural changes and create delayed job losses (as stated in its 1990 communication on La politique industrielle dans un environnement ouvert et concurrentiel). In short, States should not replace companies. 1.2.2.2. Provisions relating to public procurement Among the means of implementing an industrial policy, it is extremely common to use public procurement. In many respects, the latter plays the role of a go-between to help national industries that are in crisis (steel, textiles, construction, etc.) or considered strategic (IT, aeronautics, space, etc.), to support specific national companies, with regard to social considerations such as the employment of nationals, or to promote territorial

46

Innovation and Industrial Policies

development. The strength of this tool can be particularly important since public procurement (public procurement and concessions within the meaning of current European law) concerns all public purchasers and often represents a significant proportion of economic activity, estimated by the WTO as being in the order of 10 to 15% of the GDP, depending on the country. In France, since the 1980s–2000, public procurement has been subject to a strong legal framework that limits its ability to implement an industrial policy via this channel. To clarify this point, the rules of the GATT and the WTO will be discussed in turn, followed by those relating to European law. The GATT 1947 agreement expressly excludes public procurement from its scope. However, after having started by harmonizing customs duties and combating quantitative trade restrictions, it became clear that a real opening of markets at the global level implied in particular that States should not reserve their purchases solely to their national economic operators. This is why the signatory countries launched, with the Tokyo Round (1973–1979), a first round of negotiations on non-tariff barriers, at the end of which a series of agreements were adopted, the first of which concerned public procurement. This agreement, sometimes referred to by practitioners as the “Tokyo Round Public Procurement Code”, was signed in 1979 and entered into force on January 1, 1980. It is a multilateral reciprocity agreement applying only to countries that have accepted and signed it, aimed at opening up the field of public procurement to increased and mutual international competition through concerted, fair, transparent and non-discriminatory rules and procedures. The objective was therefore to guarantee all signatories that their products and companies will not, when public contracts are awarded to any of them, be treated less favorably than its own operators and products. Thus, signatory countries must refrain from any award subject to the supplier providing compensatory purchasing opportunities, or other similar conditions (e.g. licensing of a technology), and ensure that the words “or equivalent” appear in the technical specifications of the public contract. The agreement applies to all procedures (open, selective, one-off or over-thecounter) provided that the public procurement is carried out by a central or federal government and concerns the purchase, leasing, rental or hirepurchase of products with a value exceeding 150,000 DTSs (Special Drawing Rights: monetary instrument introduced by the IMF in 1969, the value of which is determined by a basket of currencies).

Industrial Policy and Competition

47

In addition, this agreement provided for the obligation for each signatory country to publish all matters relating to its public procurement covered by the agreement (texts, laws, etc.) and to compile statistics concerning them. It also provided for special and differential treatment for developing countries, in particular for the least developed among them. Finally, it established a Public Procurement Committee (composed of representatives of the signatory States) to make recommendations or rule to end the dispute. The Uruguay Round (1986–1994) gave rise to the WTO and a new multilateral agreement on government procurement. This time it concerns supply, service and works contracts, whether they are awarded by central (e.g. a ministry), sub-central (e.g. a region or department) or local (e.g. a municipality) entities, according to WTO terminology, with higher thresholds for the latter two types of entities. In addition, this agreement becomes evolutionary since it introduces the results of bilateral agreements signed later thanks to a system of loose-leaf sheets inserted in the initial agreement. Article XX of the agreement refers to the exceptions that signatories may apply to their own procurement when imposing and implementing: measures that are necessary to protect public morals, human, animal or plant life, or health; measures for the protection of patents, trademarks and copyrights, and the prevention of deceptive practices; or measures relating to the products of prison labour. Finally, the agreement specifies the existence of two mechanisms for redress in the event of disputes: an internal redress mechanism at the national level, and a WTO dispute settlement mechanism at the international level (see on this point the WTO website and, in particular, the Memorandum of Understanding on Rules and Procedures Governing the Settlement of WTO Disputes). The agreement currently in force since 2014 takes into account standards for the use of electronic tools, increases the fight against fraudulent practices, strengthens environmental protection and promotes the conservation of natural resources; it involves 47 WTO members. Beyond these international rules linked to the WTO which modify European law, the latter also has an extremely strong normative impact on the rules governing public procurement in France, both through treaty law and secondary legislation.

48

Innovation and Industrial Policies

European primary law, that is, the law resulting from the Treaties, has an important influence on public procurement by prohibiting measures that have an equivalent effect to quantitative restrictions and freedoms of movement, in particular with the concepts of freedom of establishment and freedom to provide services. It is on these bases that the principles of non-discrimination and publicity in public procurement have been identified. First of all, primary law remains as ordinary law and applies in the absence of secondary acts, as stated by the ECJ in its Dundalk judgment of September 22, 1988 (Case 45/87) and in its Pont sur Storebælt judgment of July 22, 1993 (Case C-243/89). It is also on the basis of this reasoning that the ECJ refused to consider it legal to reserve a percentage of public contracts for companies located in certain regions with its judgment of DuPont de Nemours Italiana of March 20, 1990 (Case C-21/88, section 13): It must be emphasized in the first place that, although not all the productions of the Member State in question benefit by comparison with products from abroad, the fact remains that all the products benefitting by the preferential system are domestic products; secondly, the fact that the restrictive effect exercised by a State measure on imports does not benefit all domestic products but only some cannot exempt the measure in question from the prohibition set out in Article 30. These elements, combined with others along the same lines, led Christine Bréchon-Moulènes (1992) to use the expression “encirclement policy” in public procurement law. This is also confirmed by the Gemeente Arnhem judgment of November 10, 1998 (Case C-360/96), paragraph 41 of which states the following: Coordinating at Community level the procedures for the award of public service contracts is to eliminate barriers to the freedom to provide services and therefore to protect the interests of economic operators established in a Member State who wish to offer goods or services to contracting authorities in another Member State. Secondary legislation in the field of public procurement is essentially based on these foundations and has been developed, on the one hand, with the establishment of fundamental principles and, on the other hand, with a

Industrial Policy and Competition

49

succession of directives. There are three fundamental principles of public procurement. First, it concerns the equal treatment of bidders, the term equality being understood both as the comparable treatment of situations that are not different and the unequal treatment of situations that are different. This principle concerns the call for competition and implies that bidders must have the same opportunities, both at the time of preparing their bids and at the time of their evaluation. Second, the principle of non-discrimination on the basis of nationality or any other criterion leading to the same result is affirmed. Thus, for all matters relating to public procurement, it is not permitted to specify a national preference clause concerning not only the operator but also supplies, equipment or services. In particular, this prevents a public purchaser from using a technical specification to foreclose European economic operators located in another Member State. Third, a principle of transparency of procedures is recognized. This implies both the publicity of the contract, which allows any company to express its interest insofar as there has been adequate publicity, and the impartiality of the award procedure, which is guaranteed by its transparency and objectivity. Secondary legislation on public procurement was also developed through a series of four legislative packages between 1970 and 2014. In the period 1980–2000, no less than nine directives were adopted. Their objective was to establish a set of harmonized rules to prepare and then extend the Single European Act, with the aim of establishing an internal market. This included extending the scope to service and network contracts, strengthening advertising and competition rules, clarifying the nature of award procedures and introducing a basis for appeals. It is also worth noting, in line with the Dundalk judgment of September 22, 1988 (case 45/87), a reference to technical standards, which would lead to the adoption in 2002 of a common vocabulary for public procurement (Common Procurement Vocabulary, CPV). Finally, this period is also crucial in that it marks with the Teckal judgment of November 18, 1999 (Case C-107/98) the advent of an exceptional approach to integrated services, also known as in house. This

50

Innovation and Industrial Policies

route, which corresponds to the case of an overlap between the contracting authority and the economic operator, thus exempts the former from its obligations in terms of public procurement, because everything happens as if the service were provided on a direct labor basis. Indeed, it is not a real contract since there is no meeting between two distinct wills. The relationship remains in the public sphere and the operator does not operate in a competitive market. In other words, the in house exception corresponds both to a lack of autonomy of the operator and to the application of the principle of freedom left to any contracting authority to satisfy its needs through its own services. Two additional conditions are required to activate this exception channel. First of all, it is an organic criterion. The contracting authority must exercise “similar control” over the economic operator to that which it has over its own services. Thus, the successful tenderer must be in a relationship of dependence with the adjudicator regarding his or her strategic choices and important management decisions. This dependence is assessed in particular by analyzing the criteria relating to the ownership of capital, the essentially non-commercial nature, the existence of a legal personality and the appointment of members of management bodies. Then there is a material criterion. This means that most of the operator’s activity is carried out with the contracting authority that holds it – or the entities that have held it since the Coditel Brabant judgment of December 13, 2008 (Case C-324/07) – in order to avoid the likelihood that the operator will enter into competition with other companies. This exception will then be extended to forms of cooperation that do not involve the existence of a dedicated operator, such as cooperation between local authorities. Today, issues relating to vertical or horizontal in house cases are governed by the 2014 directives. As a result of this legal framework for public procurement, it has gradually ceased to be one of the major tools of French industrial policy, as was the case between the end of the Second World War and the 1970s. 1.3. Conclusion Changes in economic facts and their theoretical analyses have led to the reversal of the hierarchical order that once existed between industrial policy and competition policy. The latter is now both the central policy of the European Union and the theoretical framework from which economists think

Industrial Policy and Competition

51

about the necessity (or not), and the modalities of public authorities’ interventions in the productive sphere. As we have pointed out, industrial policy appears to have crumbled away, hollowed out by competition policies. This presentation seems to us to reflect the situation in France until the beginning of the 21st Century. However, much of this development has been contested because it would prevent an understanding that the difficulties facing France would come from its gradual transition, since the end of the Trente Glorieuses, from a catching-up economy (far from the technological frontier) to an innovation and knowledge economy (designed to move the technological frontier). In other words, the suffocation of any industrial policy by competition policy would only allow us to take advantage of the best opportunities of a given situation, that is to say on the technological frontier, but not to move it. France’s industrial difficulties thus seem to come from the fact that it has not been able to find ways to take full advantage of the Third Industrial Revolution (electronics and information technology) and that it must now participate in the Fourth Industrial Revolution if it does not want to decline, which is that of artificial intelligence, robotics, the Internet of Things, autonomous vehicles, 3D printing, nanotechnologies, biotechnologies, energy storage or quantum computing, to use the fields identified by Klaus Schwab (2017). To this change in the conception of what makes French grow today, marked by theories of endogenous growth, we must also add the contributions of the new approach to the international economy, as well as the theory of clusters (or industrial clusters). If this conception is at least partially justified, it makes it possible to understand why innovation and R&D policies have occupied a growing place in France for nearly 10 years, the reasons for some of the changes in competition policies (which we have not previously developed to exaggerate and emphasize the main trend), and the renewed economic arguments in favor of an industrial policy.

2 Competition and Innovation Policy

In recent decades, most countries around the world have gradually chosen competition as the background and framework within which governments can intervene in industrial matters. This convergence is attested, at the global level, by the International Competition Network, which currently includes 336 national or regional bodies. Thus, there is almost no territory in the world today that is not directly or indirectly subject to competition rules (membership of a regional group, the WTO, or by extraterritoriality of foreign laws). These choices have in fact transformed State action by severely limiting any form of vertical intervention in order to avoid exogenous changes in production and competitive relations. This is particularly clear in the context of the European project, with the rise of European competition policy. This movement finds its main foundations in standard economic theory and more specifically in the articulation between competition, economic growth and international trade. By caricaturing, it is possible to consider that there is a pattern, identical for each country, of promoting both internal and external competition in order to obtain the maximum possible growth. In this context, the functions of public authorities are limited to allowing and encouraging business production and competitive market relations. This is also in line with the message to troubled economies, issued in particular by the IMF and the World Bank, referred to as the “Washington Consensus”, following John Williamson (1989). However, even as this liberalization movement was taking place, it did not fully capture the tectonics that could be observed in the interventions

Innovation and Industrial Policies, First Edition. Joël-Thomas Ravix and Marc Deschamps. © ISTE Ltd 2019. Published by ISTE Ltd and John Wiley & Sons, Inc.

54

Innovation and Industrial Policies

implemented in industry and innovation. Indeed, since the 1980s, developments in standard economic theory have led to a far more nuanced interpretation of the neoliberal credo that a less restrictive economic policy would always be better. The first steps in this evolution began with what became known as the “new industrial organization”, which helped to fundamentally change the clear and unambiguous liberal conclusions reached by the Chicago school in the 1970s. Thanks to game theory tools and progress in the mathematical formalization of concepts that were previously difficult to use, such as increasing returns to scale or monopolistic competition, the possibility of extending the framework of standard1 economic theory has opened up. The emergence of this new perspective quickly generated a strong interest among governments in developed countries, given the importance of the challenges and difficulties they faced, particularly in terms of growth and employment. Thus, even as these countries followed recommendations to make their economies more competitive, they also developed active innovation policies through the creation of new institutions and the implementation of horizontal interventions. In order to explain how, at the very moment when competition policies became cardinal, questions relating to innovation gradually became the central concern of public authorities, we will return first to the movement to renew the framework of thought (section 2.1), then to the advent of an innovation policy qualified in France as a “new industrial policy” (section 2.2) and finally to its ambiguities (section 2.3). We will conclude with a presentation of the PIA (Programme d’investissement d’avenir2) (section 2.4), which completes the institutional framework characterizing this innovation policy. 2.1. The renewal of the framework of thought From the end of the 19th Century onwards, economic analysis gradually consolidated around the organizational notion of equilibrium, both in terms of partial and general analysis. In this context, innovation is essentially

1 For further developments, see inter alia Bougette et al. (2015). 2 French Future Investment Program.

Competition and Innovation Policy

55

considered as a simple shift of the frontier of production possibilities that does not require any adaptation of the other parameters of the analysis. Yet economists have always been aware that innovation brings about many other structural and permanent changes in economic activities. One of the first to deal analytically with this question was Alfred Marshall who, in his Principles of Economics (1890), linked the phenomenon of increasing returns to scale with the idea of progress and therefore innovation. But, as early as the 1920s, John Harold Clapham (1922) first and then Piero Sraffa (1926) definitively demonstrated that there was an incompatibility between competitive equilibrium and increasing returns to scale. It was only through analytical developments in microeconomics in the 1970s that two solutions to overcome this obstacle emerged. The first is to use the Marshallian notion of externality, considering that returns to scale are external to firms. The second uses product differentiation with the notion of monopolistic competition, developed by Edward Hastings Chamberlin (1962 – originally published in 1933). This conceptual and analytical leap opened up new perspectives in the 1980s, simultaneously in the fields of economic growth, international trade and economic geography, thus profoundly renewing the analytical treatment of innovation. More specifically, the renewal of the framework has completely changed the way economic policy recommendations are conceived, not only by redefining the determinants of imperfect competition (section 2.1.1), but also by placing innovation at the heart of growth (section 2.1.2), while opening up issues relating to the geographical location of innovation and production activities (section 2.1.3). This new concept thus underpinned the main economic policy guidelines established by the European Union since 2000, known as the Lisbon Strategy (section 2.1.4). 2.1.1. A new competitive economics Faced with the essentially pessimistic message of classical authors regarding the possibility of stable and permanent growth, Robert Solow’s neoclassical analysis (Solow 1956, 1957) offers significant intellectual comfort. Indeed, the latter believes that, thanks to the competitive functioning of markets and the substitutability of factors of production, there are forces that will eliminate the instability and disequilibrium previously highlighted in the work of Roy Harrod (1939) and Evsey Domard (1947).

56

Innovation and Industrial Policies

Stable and sustainable growth is therefore possible, along a balanced growth path, if markets operate in a competitive way. The social optimum and the competitive equilibrium always coincide there. Moreover, within this framework, there is a phenomenon of absolute convergence of economies towards an identical rate and growth path for all economies with the same demographic, technological and institutional characteristics (i.e. belonging to the same convergence club). The modeling proposed by Solow and its extensions would, until the 1980s, constitute the core of economic growth theories and therefore served as a basis for the growth and macroeconomic stabilization policies of industrialized countries. However, this analysis has some limitations. The main one is that, even though it agrees with the idea that technological progress is the main determinant of growth, as Adam Smith already noted as early as 1776, and that it makes it possible for the first time to deal with technological progress in a formalized analytical framework, it continues to give only a preliminary explanation. In a way, as Gilles Saint-Paul (1996) notes, technological progress appears to be a “black box”. Admittedly, Robert Solow demonstrates that capital accumulation is not sufficient to explain growth and that technological progress must necessarily be introduced, but the latter, because of modeling constraints (i.e. the decreasing returns on accumulated capital), appears either as totally exogenous (“fallen from the sky”) or as applying only to the new goods in which it is incorporated. Thus, economists do not have, within this framework, the knowledge of the elements and mechanisms that determine technological progress or its evolution. By forcing the line very hard, they do not know how the State can pursue a growth policy with major effects since they do not know how to influence its main driving force, technological progress, whose contribution to growth appears to be much greater than that of physical capital since it oscillates, according to Dominique Guellec (2009, p. 59), between 50 and 80%. However, the problem was all the more urgent since, from the mid1970s, on the one hand, the United States was facing productivity difficulties and, on the other hand, the growth of European countries, corresponding to the period of the Trente Glorieuses, and that of some Asian countries, which began in the 1960s, gradually seemed to be coming to an end. By breaking the technical barriers that prevented growth modeling from progressing, analytical developments in industrial organization made it possible to identify the factors that determine technological progress. This is

Competition and Innovation Policy

57

the origin of the new growth theories, which are therefore generally referred to as “endogenous growth theories”, since they make it possible to explain and analytically treat technological progress in models. These models propose microeconomic mechanisms within a general equilibrium framework. In these approaches, technological change has its origin in the investments of economic agents in basic research, R&D, education, learning by doing, and physical capital. The common point of these approaches is the existence of increasing returns of scale in the investment. The characteristic feature here is that there are no diminishing marginal returns in terms of investment, which allows an accumulation process, and therefore growth. It then comes from an endogenous creation and not from exogenous advantages. Various justifications have been put forward. Thus, in his seminal model, Paul Romer (1986) considers that capital today is largely immaterial and that it corresponds much more to ideas than to objects. He then proposes a model where, in a competitive environment, the marginal return on capital remains constant with firms deciding on investments in physical capital or knowledge; but where all of these investments result in a positive externality linked to the creation of additional non-rivaled knowledge (i.e. external economies of scale through learning by doing), which is the source of technological progress. The origin of growth thus lies in the stock of knowledge, assimilated to the stock of capital, since a worker’s efficiency depends on the total stock of knowledge acquired by other workers. In a later model, hinging upon three sectors (research, intermediate good and final good), Paul Romer (1990) identifies research as the place of externality and the source of technological progress. The growth rate then depends on the pace of technological progress, which is the result of the number of agents engaged in research, the latter being the source of increasing returns. It is the pursuit of profit that generates new ideas. It should then be noted that the growth rate that an economy can achieve is not the one that would be socially optimal, since agents do not take into account the positive externality that their private investments generate: they underinvest compared to the socially optimal level. This situation naturally opens up the possibility of State intervention, which may, in particular through a system of subsidies, encourage companies to carry out more research. But the State must also take into

58

Innovation and Industrial Policies

account that companies will only be encouraged to innovate to the extent that they can have a profit (i.e. a price above marginal cost), which necessarily has an influence on the competition policies that the State must pursue. Another approach was developed by Robert E. Lucas (1988) who, inspired by Gary Becker, proposed two versions of an endogenous growth model based on the accumulation of human capital. In the first, employees decide to devote part of their working time to improving their skills through training and it is this accumulation of human capital that generates an externality. In the second case, this accumulation comes from a learning-bydoing effect and is therefore proportional to production. Each time, the use of knowledge does not reduce its effectiveness and the acquisition of human capital is all the more valued by an agent when the others have a high level of human capital. Thus, all other things being equal, the higher the average level of knowledge in an economy, the higher the productivity of each of its firms will be. In a different vein, some models, such as Robert J. Barro’s (1990), make public capital the source of a positive externality that benefits all private sector firms and prevents private capital from having decreasing returns to scale. The model then determines an optimal tax rate that allows growth. Two effects must be taken into account: on the one hand, public spending generates a positive externality, but on the other hand, since it comes from taxes, it reduces the profitability of investments, which reduces capital accumulation by companies. Finally, other models consider technological progress as the result of voluntary R&D activity and propose a formalization to explain its origin and rationality for agents. The causality is then clear: research leads to innovation and innovation leads to technological progress. The distinction is generally made between models of product range expansion (horizontal product differentiation à la Hotelling) and those based on product quality improvement (vertical differentiation à la Chamberlin). Among the first is one of the models developed by Gene Grossman and Elhanan Helpman (1991) in which research leads to innovation by producing new consumer goods. As consumers like diversity, they have a greater utility since they will be able to spread their consumption over a greater number of goods. Real incomes therefore increase as a result of these innovations (Acemoglu 2009).

Competition and Innovation Policy

59

Concerning the second type of model, we can look at that of Philippe Aghion and Peter Howitt (1992) who, taking up Schumpeter’s developments around the notion of “creative destruction”, focus on the idea that innovation corresponds to an improvement in the quality of inputs by a replacement effect, that is new inputs replace the previous ones. In each of these cases, technological progress results from competition between innovative firms. From this brief overview of some of the main contributions to endogenous growth, which is further summarized in the work of Philippe Aghion and Peter Howitt (2008) or Jean-Luc Gaffard (2011), it appears that the State can, in some cases, intervene in a beneficial way, but in new forms. Thus, public expenditure on R&D, training, infrastructure, or savings support can again be justified in terms of its effect on technological progress. But these are no longer vertical policies decided directly by the State, in the form of major programs or direct aid to companies. From now on, the State must promote positive market failures (externality and public good) through horizontal policies, as well as institutional reforms, making it possible to coordinate and thus multiply the socially positive effects of agents’ decisions. The main challenge of public policies then becomes that of promoting innovation, the engine of economic growth, in order to achieve the socially optimal equilibrium. By way of illustration, two proposals for the implementation of such policies for non-targeted subsidies were particularly noted (Aghion and Howitt 1997). The first is that of Paul Romer (1993) who suggested that this type of subsidy should be distributed to companies, according to the quality of the projects they propose, by commissions composed of industrialists from the sectors concerned. The second is from Michael Kremer (1998) who proposed to use a system designed to have patents repurchased by the State to put them in the public domain, in the same way as the French government did in the 19th Century with the Daguerreotype photographic process. In its system, the State informs itself of the value of the innovation concerned by using an auction mechanism and then proposing a sum corresponding to the maximum auction (revealing the private value) plus a margin (corresponding to the difference between the social value and the private value), in order to encourage innovation. These new forms of public policy are all the more important because some models of endogenous growth highlight hysteresis phenomena, and

60

Innovation and Industrial Policies

therefore irreversibility, which means that history matters and that it can be difficult to rid oneself of a bad start. 2.1.2. The geography of innovation In addition to the new theories of growth, the renewal of microeconomics has also allowed the simultaneous emergence of new theories of international trade, then of the new economic geography. In both cases, Paul Krugman’s work is at the heart of these developments. By retaining the notions of oligopoly, increasing returns to scale and product differentiation, the new theories of international trade move away from traditional models with constant returns and perfect competition. They make it possible to better reflect, on a theoretical level, the fact that international trade is mainly between advanced and similar-sized countries, that it is composed of a high share of intra-industry trade (cross-trade in similar products), and that less developed countries do not necessarily converge towards advanced countries. These new theories are also the basis for a review of the effects of protectionism and what is known as “strategic trade policy”. We will briefly come back to these different points. Before the 1980s, formalized models of international trade explained that trade between countries could only take place to the extent that they differed in their technologies or factor endowments. These are the two main determinants of trade illustrated respectively by the Ricardian model (heterogeneity of technologies) and the Heckscher-Ohlin model (heterogeneity of allocations in factors of production). However, new theories of international trade demonstrate that in an imperfect competitive environment, economies of scale can explain intraindustry trade. This is what Paul Krugman (1979, 1980) does by demonstrating that this type of trade comes solely from the taste for consumer diversity and business differentiation strategies. Taking up some of this work and the contribution of James Brander (1981), Brander and Krugman’s (1983) model shows that intra-industry trade can also be explained by the strategic interaction of corporate behavior. However, considering that “increasing returns are as fundamental a cause of international trade as comparative advantage” (Krugman 1987, p. 133)

Competition and Innovation Policy

61

inevitably lead to a distinction between this new approach to international trade, based on the concepts of imperfect competition and increasing returns, and the old one, based on those of perfect competition and comparative advantage. This new orientation raises an additional question since “these new models call into doubt the extent to which actual trade can be explained by comparative advantage” (ibid., p. 131). In other words, it becomes impossible to predict the direction of intra-industry trade flows based on country characteristics (Helpman and Krugman 1985). This approach therefore leaves “a great deal to the arbitrariness or chance of history in the origin of specialization” (Laussel and Montet 1989, p. 7). This questioning of the principle of comparative advantage also leads to a rethinking of the idea that free trade is always the optimal trade policy. In this perspective, Barbara Spencer and James Brander (1983, 1985) show that it may be better for a state to intervene in industrial matters, for example by financing R&D, subsidizing exports or limiting imports, that is by pursuing a strategic trade policy. This idea of protecting the domestic market in order to promote exports is also reflected in the conclusions of Krugman’s model (1984) where, because of the presence of economies of scale, it can be a winning situation for a country to protect its national champion. Although it is impossible, in a situation of imperfect competition, to demonstrate that free trade would necessarily be the best trade policy, a large majority of economists, including Paul Krugman, believe that: free trade is not passé, but it is an idea that has irretrievably lost its innocence. Its status has shifted from optimum to reasonable rule of thumb. There is still a case for free trade as a good policy, and as a useful target in the practical world of politics, but it can never again be asserted as the policy that economic theory tells us is always right (Krugman 1987, p. 132). In a subsequent contribution, he extends his observation by relativizing the scope of strategic trade policy: After several years of theoretical and empirical investigation, however, it has become clear that the strategic trade argument, while ingenious, is probably of minor real importance (Krugman, 1993, p. 363).

62

Innovation and Industrial Policies

Apart from their intrinsic interest, it is noteworthy that the analyses developed by the new theories of international trade are partly in line with those of endogenous growth theories. For example, Robert Lucas’ (1988) model shows that if a country has high externalities of learning, it may not have an interest in participating in international trade. Indeed, this country could be led to specialize in industries with low externalities of learning, which would only allow it a low growth potential. In other words, the traditional conclusions of international trade are based on static efficiency but do not take into account dynamic efficiency. For such a country, the solution would then be to close its borders, while agents invest in sectors with high external learning externalities, and then open them again. In contrast, Gene Grossman and Elhanan Helpman (1990) show that international trade promotes growth by allowing advanced countries to devote more resources to innovation by importing the most common goods from least developed countries. They then mimic the new goods resulting from innovation in advanced countries, but global growth is stronger than in the absence of trade between countries. In the early 1990s, reasoning at the national level with the monopolistic competition model also led Paul Krugman to renew the analysis of spatial agglomeration issues of productive activities within regions and cities. As Matthieu Crozet (2009) points out, this transition is a natural one since, on the one hand, the new theories of international trade emphasize the role of the spatial dimension by providing a theoretical basis for gravity equations (i.e. the importance of bilateral trade flows increases with the size of the countries involved and decreases with the geographical distance between them), and, on the other hand, these theories support the existence of a link between the size of national markets and exports. Indeed, when there are transport costs, the more a country has increasing returns due to a large domestic market, the more differentiated goods it will produce and the more it will export them, to the detriment of homogeneous goods (this is the home market effect). In other words, it is a relationship “more than proportional between a country’s share of the world production of a good produced with increasing returns, and that country’s share of the world demand for that good” (Crozet 2009, p. 524). This effect thus provides an explanation of the phenomena of geographical agglomeration of production between countries and within a country close to the main market, when there are increasing transport costs and returns. The further away from the bulk of consumers, the less specialized we will become in the production of goods with increasing returns.

Competition and Innovation Policy

63

Continuing along this path, Krugman’s (1991) model endogenously explains how dispersion forces (linked to transport costs) and agglomeration forces (linked to market size and the existence of fixed costs, and therefore increasing returns to scale) are articulated. It is a model with two countries each with two sectors. The first of them is in perfect competition, and the second in monopolistic competition with companies benefiting from internal economies of scale. In the latter, workers are mobile and the good incurs an iceberg type transport cost (i.e. the transport cost is paid in units of the good transported). The agglomeration in this model is then based on two financial externalities: – companies in sectors with increasing returns are locating themselves in the main market; – workers in these sectors are also moving to this country because they are better paid and prices are lower because there is no transportation cost. These two externalities will reinforce each other and create a selfsustaining process of agglomeration. Thanks to this model, Paul Krugman manages to integrate regional and spatial economic analyses into a general equilibrium framework and to bring out the location of productive activities through market dynamics. It thus revitalizes areas that were once relatively peripheral in relation to the central corpus of the discipline. This canonical model, also known as “centerperiphery”, is the birth certificate of the “new economic geography”. Empirically, the latter is all the more important because, as a result of European economic integration through the free movement of persons, goods and capital, the European Union no longer seems understandable with the classic tools of international trade theory. The consequences of work on the new economic geography on economic policy debates are major, particularly at the level of the European Union. Indeed, while it was accepted, for models with constant returns, that the formation of a free trade area was favorable to all the countries in the area, this new literature leads to a less clear conclusion. It is indeed possible that, although collectively optimal, a process of agglomeration towards the countries of the center may be set up at the expense of the countries on the periphery. By way of illustration, these models develop the conclusion that transport infrastructure can, under certain conditions, lead to an aspiration of productive activities from outlying regions to central regions, thus

64

Innovation and Industrial Policies

contributing to the deindustrialization of the former. Using literature on endogenous growth, Paul Krugman (1995) manages in particular to highlight the existence of an arbitration between growth and spatial cohesion. This result leads, for example, to questions on the relevance of the European Union’s regional policies, developed in particular through the Structural Funds (about one third of its budget), which aim at the convergence of GDP per capita levels in European regions. These models also highlight the existence of multiple balances with path dependency, in other words historical accidents can lead to permanent effects due, in particular, to self-reinforcement mechanisms. And once companies have set up and invested in a location, economies of scale will lead them to stay there. With regard to innovation, empirical studies show that spillover effects are generally localized. Thus, Jaffe et al. (1993) highlight the importance of geography by demonstrating the existence of a significant relationship between the geographical origin of patents filed and the geographical origin of patents cited by them. This strong spatial polarization of innovationrelated activities would come from the fact that its impact would be all the more important, given the strong geographical proximity. This relationship also corresponds to the results of research combining economic geography and endogenous growth, as shown in the synthesis proposed by Richard Baldwin and Philippe Martin (2004). The latter emphasizes the major role played by knowledge externalities in the spatial location of R&D activities. Laboratories and companies are encouraged to group geographically to take full advantage of tacit transfers of knowledge and skills, through interpersonal relationships and local seminars, for example. The analysis of the relocations of French companies and the model developed by Gilles Duranton and Diego Puga (Duranton and Puga 2000, 2001) reinforce this approach by demonstrating that firms locate themselves differently. They prefer diversified areas when they are in the process of experimenting with their activities, in order to be able to benefit from externalities in the style of Jacobs (1969), also known as urbanization, that is externalities resulting from the grouping in the same place of companies belonging to different sectors. On the contrary, they move to specialized areas when they have chosen their activities in order to benefit from Marshallian externalities, also known as specialization, that is externalities

Competition and Innovation Policy

65

resulting from the grouping in the same place of firms belonging to the same sector. In other words, the analysis proposed by Alfred Marshall (1890), concerning industrial districts, provides only a partial understanding of localization phenomena. In terms of innovation, which is the new core of productive activities, the geographical dimension is more complex because it requires other forms of proximity as highlighted by the advent of the concepts of cluster and ecosystem. Previous analyses have largely disrupted the certainties that previously prevailed and renewed the thinking frameworks of economic analysis. This renewal has logically led to the emergence of new forms of industrial policies, focused on innovation, as evidenced in particular by the European choices of the early 21st Century. 2.1.3. Innovation and competition policy: the Lisbon strategy Meeting in Lisbon in March 2000, the Heads of Government of the then 15 Member States of the European Union took the decision to make Europe, by 2010, “the most competitive and dynamic knowledge-based economy in the world, capable of sustainable economic growth with more and better jobs and greater social cohesion”. This ambition, known as the Lisbon Strategy, has become the main thrust of the European Union’s economic policy. Its main characteristic was to be fully in line with the new principles of endogenous growth theory. Indeed, the justification for this strategy was that, in the field of research and especially innovation, Europe was lagging significantly behind the United States, which explained its poor growth performance. However, the main consequence of this delay was an increase in unemployment, a decrease in productivity gains and a slowdown in the standard of living. The same principles of endogenous growth have therefore also been mobilized to face these difficulties. Thus, the Lisbon strategy initially had two components: on the one hand, an economic component aimed at promoting the transition to a knowledge-based economy based on research and innovation; on the other hand, a more social component aimed not only at investing in education and training to implement an active employment policy, but also at modernizing social protection to promote the fight against exclusion. Then, a third environmental component was added to the two previous ones in 2001, at the Göteborg European Council, which

66

Innovation and Industrial Policies

added the environmental dimension to the strategy agreed in Lisbon, by approving a sustainable development strategy. The objective of this third component was to address climate change by proposing to decouple economic growth from the use of natural resources, with the aim of promoting sustainable development. To carry out a mid-term review of the Lisbon Strategy, the European Commission mandated a “high-level group” composed of a small number of highly qualified people under the chairmanship of the Dutchman Wim Kok (Prime Minister of the Netherlands from 1992 to 2002). The findings of this commission were made public in November 2004, under the name of the Wim Kok report. As the latter reminds us: […] Lisbon aims to raise private and public research and development spending as the centerpiece of a concerted effort to increase the creation and diffusion of scientific, technological and intellectual capital. It aims to foster trade and competition by completing the single market and opening up hitherto sheltered and protected sectors. It aims to improve the climate for business and enterprise. It aims to secure more flexibility and adaptability in the labor market by raising educational and skill levels, pursuing active labor market policies, and encouraging Europe’s welfare states to help the growth of employment and productivity rather than hinder it. And it aims for growth to be environmentally sustainable (Wim Kok 2004, p. 8). At the same time, however, the Wim Kok report notes that “but halfway to 2010 the overall picture is very mixed and much needs to be done in order to prevent Lisbon from becoming a synonym for missed objectives and failed promises” (ibid., p. 10). Nevertheless, the Report considers that “[…] despite disappointments Lisbon is not a picture of unrelieved gloom, as some like to paint. There has been significant progress in employment between the mid1990s and 2003” (ibid.). However, this report concludes that the Lisbon strategy has failed: “Clearly there are no grounds for complacency. Too many targets will be seriously missed. Europe has lost ground to both the US and Asia” (ibid., p. 11). But this failure cannot be the responsibility of the strategy itself, since it stems from what should be considered the right economic theory. Indeed:

Competition and Innovation Policy

67

The evidence is overwhelming that the higher research and development expenditure, the higher subsequent productivity growth. One of the preconditions for any increase in European productivity growth is to raise R&D spending (ibid., p. 19). The causes of failure can therefore only be exogenous, for example the fact that: The ink had scarcely dried on the agreement before the worldwide stock market bubble imploded, the epicenter of which was the collapse of the overvalued prices of American dot.com and telecom shares amid evidence of financial and corporate malpractice. Skepticism mounted about the potential of the knowledge economy (ibid., p. 9). Similarly, the Wim Kok report notes that “the terrorist attacks on the US on 11 September 2001 and subsequent events further darkened the international climate” (ibid. p.10), in addition to other events such as “[…] tensions between Europe and the United States [which] have resulted in some bitter trade disputes” (ibid.); and the increase in oil prices due to increased demand and supply insecurity, “[…] both dampening current economic activity and lowering forecasts for the immediate future” (ibid.). The combined effects of all these events would therefore explain why, over the period 2000–2004, “[…] the overall performance of the European economy has been disappointing” (ibid.). However, other reasons were also mentioned, in particular those put forward by Élie Cohen, who considers that “the Lisbon strategy has suffered above all from a lack of will and embodiment insofar as it was not supported by anyone, by any authority” and that it “achieved a reconciliation in terms of partially contradictory objectives (growth, employment, sustainable development, preservation of the social model). It was a strategy without a general, without weapons and without troops” (Cohen 2012, p. 131). The recommendations of the Wim Kok report are therefore in line with the continuation and pursuit of the Lisbon Strategy, since they suggest supporting research by stimulating scientific excellence, adapting education and training systems to the knowledge society and pursuing the completion of the European market by facilitating the free movement of persons, goods, services and capital within an area without internal borders. Above all,

68

Innovation and Industrial Policies

however, the Kok report considers it essential to pursue the same liberal strategy by “identifying and removing barriers to competition, [...] in order to foster further liberalization and open more sectors to EU-wide competition” (ibid., p. 25). Following the Wim Kok report and the various reactions it generated, the European Commission proposed in 2005 a European Partnership for Growth and Jobs, which focuses the objectives of the Lisbon Strategy on growth and jobs. This choice does not imply abandoning the other objectives of social cohesion and the environment because the Commission considers, in accordance with its liberal conception, that the achievement of the latter is based above all on growth and employment. The main innovation regarding the content of this new strategy is the consideration of the industrial dimension. In particular, the Commission proposes to reform State aid to support research and innovation, mainly in favor of SMEs. It also indicates that it will support and promote innovation poles, which aim to combine research and industry. It is then possible to make the following observation: Whereas in 2000, the Lisbon Strategy placed a strong emphasis on information and communication technologies, which can probably be explained by the boom they experienced at that time, it neglected the industrial dimension of the European economy. The Kok report does not refer to it (Delebarre and Garrigue 2005, pp. 41–42). Also, in 2002 and 2004, the European Commission published communications setting out the main orientations envisaged for industrial policy. In particular, it affirmed the importance of industry in the European economy and highlighted the structural changes affecting it. The Commission has therefore defined three areas to support these changes and strengthen competitiveness. The first was on improving the regulatory framework, the second on developing synergies between Community policies and the third on developing the sectoral dimension. However, as also indicated in the Delebarre and Garrigue report: In terms of methodology, the implementation of an EU industrial policy is difficult because it is cross-cutting and involves many policies (competition, internal market, research, training) (ibid., p. 42).

Competition and Innovation Policy

69

However, it is this problem of method that the French public authorities have tried to solve by developing a real innovation policy under the title of “new industrial policy”. 2.2. Innovation policy as a “new industrial policy” The slowdown in economic growth and the relocation movements observed in France since the early 1990s, but also the transition to a globalized economy characterized by the emergence of new competitors, in particular China and India, have led to a renewed reflection on the place of industry in the global economic environment. This reflection marks an important turning point since it heralds a return to the idea of industrial policy, even though it had been excluded from political discourse for some twenty years. Above all, it is accompanied by a change of perspective, perfectly diagnosed in 2000, in the report produced by Élie Cohen and Jean-Hervé Lorenzi, entitled Politiques industrielles pour l’Europe. These two authors note that there is indeed “some paradox in wanting to revoke European industrial policies” (Cohen and Lorenzi 2000, p. 9). Indeed: A broad agreement has been reached over the last ten years on the occasion of the forced march towards the euro for competitive disinflation policies, for the abandonment of sectoral policies, for the adoption of competitive environmental policies. Moreover, in a rare movement of political and administrative coherence, the Ministry of Industry has disappeared, the Treasury Directorate has disinvested and closed the specialized counters, and despite the changes, the same disengagement policy has been implemented (ibid., p. 10). From the outset, the question was asked at European level because the answer actually came from Europe with the Lisbon Strategy. It is in this context, but also in part to respond to the failures of this strategy, that France has developed a “new industrial policy” articulated in three complementary components: pôle de compétitivité (competitiveness clusters) (section 2.2.1), a new institutional framework to foster innovation (section 2.2.2) and a set of support measures for innovative SMEs (section 2.2.3).

70

Innovation and Industrial Policies

2.2.1. Innovation and territory: competitiveness clusters It was in the 2000s that a real innovation policy was gradually introduced. It has the particularity of being linked to a new spatial planning policy which aims to give an essential role to the regions. As Jean-Sébastien Scandella points out, this story “officially began on 13 December 2002 at an Inter-ministerial Council for Spatial Planning and Development (CIADT), with the presentation by Prime Minister Jean-Pierre Raffarin of the foundations of a new French spatial planning policy” (Scandella 2008, p. 11), whose objective was to adapt the French economy to the Lisbon strategy. In its 2003 report to the Prime Minister, entitled Une nouvelle politique de développement des territoires pour la France, the DATAR states that “the novelty has just been that metropolises are no longer considered solely in terms of urban planning and development, but also in terms of the role they play, alongside the regions, in the development of economic poles of excellence and in attracting companies and visitor flows” (DATAR 2003, p. 7). This first mention of the notion of an economic center of excellence will gradually evolve over the course of the report towards that of a competitiveness center, after having adopted the concepts of a technological, industrial or scientific center of excellence in turn. The following year, in the DATAR report entitled La France puissance industrielle: une nouvelle politique industrielle par les territoires, the concept of a competitiveness cluster appeared to be perfectly defined and stabilized: Competitiveness clusters, understood as the articulation of the principle of cooperation and innovation, constitute a response aimed at placing companies located in the national territory in a position of performance and making the territories a factor of competitiveness and attractiveness of the French economy (DATAR 2004, p. 42). The DATAR thus proposed the implementation of a competitiveness cluster policy aimed mainly at strengthening, in the territories, the links between industry and research, and industry and education, but also at stimulating cooperation in the field of innovation by promoting relations between companies, research laboratories, both private and public, universities and grandes écoles. The originality of the approach was to mobilize new developments in economic analysis by explicitly relying on:

Competition and Innovation Policy

71

the emergence of the knowledge economy, which strongly extends the information economy, and the search for the minimization of transaction costs, which has been one of the foundations of the most traditional economic analysis. [...] The knowledge economy is transforming knowledge from a factor of production to a production in its own right that makes innovation, not a stage in the accumulation of a stock of values, but a continuous process that determines competitive development (ibid., p. 70). The same articulation between territory and innovation is at the heart of the report commissioned in December 2003 from Christian Blanc by the Prime Minister to define the conditions for setting up competitiveness clusters. Submitted in April 2004, this report entitled Pour un écosystème de la croissance proposes to move from an imitative economy to an innovation economy involving a strengthening of decentralization and regional dynamics, an increase in research funding and a rebuilding of the university system. This new industrial policy, based on the principle of competitiveness clusters, was officially initiated by the CIADT3 on September 14, 2004. Since that date, a competitiveness cluster has meant “the combination, in a given territory, of companies, training centers and research units engaged in a partnership approach designed to generate synergies around joint projects of an innovative nature, and having the critical mass necessary for international visibility” (CIADT of September 14, 2004, p. 5). The procedure for setting up competitiveness clusters was carried out in three stages: first, a call for tenders launched in November 2004, then the receipt of responses in February 2005 and finally selection and validation by the CIADT in July 2005. It should be noted that a decree of October 13, 2005 broadens CIADT’s missions by transforming it into CIACT (French Interministerial Committee for Spatial Planning and Competitiveness),

3 Created in 1960, the CIADT or Inter-ministerial Committee for Spatial Planning and Development decides on long-term perspectives and choices in terms of spatial planning and the use of funds from the various funds within its field of competence and arbitrates the financial commitments of the ministerial departments.

72

Innovation and Industrial Policies

which is thus responsible for deciding on the objectives of the national policy of territorial attractiveness, competitiveness and cohesion. While the pre-selection committee had recommended a reduced number of poles, the concern to irrigate the territory rather than favor a few large poles prompted the government to increase the number by classifying them into three categories. Thus, in July 2005, out of 105 projects submitted, CIADT approved 67 clusters (list reduced to 66 after the merger of two projects), then five new clusters in July 2007. These 71 competitiveness clusters are divided into seven global clusters, 10 with a global vocation and 54 with a national vocation. It should be noted that the categorization between “global”, “global vocation” and “national” poles was abandoned for the third phase of pole deployment and that the first phase lasted from 2005 to 2008, the second from 2009 to 2012, the third from 2013 to 2017, while the fourth phase was launched in June 2018. Nearly one billion euros of public funds have been invested since the launch in 2005, out of a total of 1.5 billion euros planned until 2008. These poles were in principle selected on the basis of three criteria: the existence of a science-industry partnership, reference to a market and/or a technological field, and the requirement of a partnership mode of governance. Their implementation is managed by dedicated coordination committees, organized into a funder committee and a scientific committee that is more specifically in charge of monitoring and technical evaluation of the cluster’s results, as well as by a correspondent of the Inter-ministerial Working Group (GTI). The GTI is, for its part, responsible for a general mission of supporting and monitoring the labeled clusters. It is also responsible for overseeing the labeling process for new clusters that may emerge in the future. In the same spirit, but with a different perspective, the Conseil Régional de Provence-Alpes-Côte d’Azur created the Pôles régionaux d’innovation et de développement économique solidaire (PRIDES) at the end of 2006, the vocation of which was to encompass innovation in the broad sense and solidarity to contribute to the creation of wealth and sustainable jobs by taking into account social and environmental responsibility issues.

Competition and Innovation Policy

73

Figure 2.1. Map of the 71 competitiveness clusters in 2007 (source: DGCIS/DATAR). For a color version of this figure, see www.iste.co.uk/ravix/innovation.zip

This PRIDE policy is part of an approach equivalent to that already undertaken in favor of competitiveness clusters. This is what makes Jean-Sébastien Scandella say that this type of approach “overlaps with the national initiative, a situation that can sometimes create some confusion” (Scandella 2008, p. 45). However, the objective of this policy is different since it involves concentrating the region’s means of intervention towards regional companies engaged in a cooperative approach around a global economic development strategy. More precisely, it is a question of densifying the productive fabric of SMEs and ensuring balanced regional development in the face of the movement towards metropolization and concentration of wealth; but also of improving the attractiveness of the region by affirming strong economic skills (Enjolras 2010).

74

Innovation and Industrial Policies

Figure 2.2. Map of the 67 competitiveness clusters in 2018 (source: DGCIS/DATAR)

This new approach to public intervention in industrial matters, based on a link between innovation and territory, undoubtedly marks a break with previous policies. It will be symbolically completed with the disappearance of the Commissariat general du Plan, replaced in March 2006 by the Centre d’analyse stratégique. Beyond the symbol, this institutional transformation is partly the consequence of the harsh observation expressed in Christian Blanc’s report: At the national level, our country’s economy is frozen in an organization that dates back to the primary era: that of the reconstruction of the country and of the Trente Glorieuses. Our actors are fossilized in vertical systems that make the

Competition and Innovation Policy

75

interactions between research, teaching and business lose all the vitality from which innovation and competitiveness are born (Blanc 2004, p. 1). In particular, it reflects a radical change in the way the State acts in the industrial field, since it endorses the abandonment of a centralized approach, based on the principle of direct intervention by public authorities which is in turn based on a link between public research, public procurement and public companies, in favor of a new, more decentralized and incentive-based approach, based on the principle of a public-private partnership. However, this new system for promoting innovation, which is based on the principle of networking, in the same territory, a plurality of actors (SMEs, large companies, research centers, training centers) around innovative projects, does not in itself present a real originality. Indeed, even if it differs in several aspects, the idea of a competitiveness cluster is an extension of the concepts of industrial districts, clusters or local productive systems (Weil and Fen Chong 2008) and refers more generally to the concept of an ecosystem of innovation and production (Guilhon 2017). It should also be noted that it is not very different from the European concept of “innovation poles” which the Community Framework for State Aid for Research and Development and Innovation defines as: independent undertakings—innovative start-ups, small, medium and large undertakings as well as research organisations— operating in a particular sector and region and designed to stimulate innovative activity by promoting intensive interactions, sharing of facilities and exchange of knowledge and expertise and by contributing effectively to technology transfer, networking and information dissemination among the undertakings in the cluster4. The approach and measures resulting from Christian Blanc’s report were reinforced by those of Jean-Louis Beffa’s report, Pour une nouvelle politique industrielle (2005), commissioned by President Jacques Chirac in September 2004. Indeed, in its mission letter, it was asked in particular to “ensure the coherence of its proposals with the public policies adopted in the industrial sector, in particular the innovation plan, the attractiveness plan, the 4 Journal officiel de l’Union européenne, 2006/C 323 of December 30, 2006, p. 10.

76

Innovation and Industrial Policies

competitiveness clusters and the law on the orientation and programming of research, which is currently being drafted, as well as with the actions carried out at European level” (Beffa 2005, pp. 5–6). It was therefore a new institutional framework that was being sought in order to guarantee the coherence and relevance of the general innovation support mechanism constituted by competitiveness clusters. 2.2.2. A new institutional framework for innovation The new philosophy that is supposed to drive this new industrial policy is clearly stated in Jean-Louis Beffa’s report: In France, the redefinition of an industrial policy means giving new meaning to the missions of prospecting, coordination and encouragement [...] The renewal of industrial policy must be organized around the promotion by the State of long-term industrial technological programs. This action must be carried out as closely as possible to ‘pre-competitive’ industrial development and in a way that is complementary to the public effort that must be directed towards fundamental research. This approach is based on a partnership between private companies and public authorities (Beffa 2005, p. 8). The new institutional framework established on the basis of the recommendations of the Beffa report has two pillars. The first pillar consists in creating a new financing and information system to promote innovation. This system is based on three “resources agencies” whose function is to select, finance and support innovative projects. The first agency is the OSEO group, created in 2005 by the merger of ANVAR and BDPME (Banque du développement des PME). It supports national and regional policies to support and assist SMEs (through advances, loans or guarantees) in the various phases of their life cycle (creation, innovation, development, transmission). It thus facilitates access for project leaders and entrepreneurs to financing from banking partners and equity institutions. The second is the Agence nationale pour la recherche (ANR). Dedicated to fundamental and applied research projects, it must also contribute to the

Competition and Innovation Policy

77

transfer of public research results to companies. Created in April 2006, the ANR is responsible for selecting research projects through calls for projects, after they have been put out to competition. For 2007, it has a commitment capacity of €825 million for projects with a maximum duration of four years. These can be proposed by public bodies or companies and one of the Agency’s objectives is to promote partnerships between these two types of actors. It is responsible for awarding Carnot labels to public research centers that have demonstrated their ability to cooperate with companies. The first 20 labels were awarded in March 2006, with funding of €40 million. The third is the Agence de l’innovation industrielle (AII), which was created on August 25, 2005, in accordance with the conclusions of the Beffa report. The AII is an industrial and commercial public institution whose mission is to support major innovation projects (requiring high investments, several tens of millions of euros), based on initiatives by companies. The government has set a target for SME participation in the IIA of 25% of programs, either in partnership or subcontracted. The IIA financial support takes the form of grants or repayable advances, depending on the development phases of the research. Its aim is to redirect industry towards the high-tech sector, by helping large companies to carry out research programs that go beyond their usual R&D and that direct them, through the technological breakthrough produced, towards promising markets of global dimension. The IIA has a unique mode of intervention: the Programme mobilisateur pour l’innovation industrielle (MIPI) (Mobilization Program for Industrial Innovation). The financing mechanism is supplemented by the Fonds unique interministériel (FUI), which is responsible for financing “collaborative” research and development projects of competitiveness clusters. It is part of the Business Competitiveness Fund of the Ministry of Finance, Industry and Employment. It provides subsidies to companies (SMEs or large companies) and research laboratories. To benefit from this support, R&D projects must be labeled by a cluster and involve at least two companies and a laboratory or research center. The size of the projects varies from €1 to €10 million. The second pillar is represented by the Programmes mobilisateurs pour l’innovation industrielle (PMII), directly inspired by the Beffa report. A PMII is an industrial program that is carried out by a company, usually in partnership with research institutions, but also with other companies.

78

Innovation and Industrial Policies

Four criteria characterize the technological and industrial dimensions of a PMII: – a large European and global market of significant size and the objective of taking a significant share; – a strong technological innovation, involving R&D costs of one to several tens of millions of euros; – the leading role and financial involvement of a leading industrialist, responsible for implementing the program and coordinating a cooperative system including public research laboratories, other partner companies, and possibly major clients and prescribers; – a medium-term horizon, from 5 to 15 years depending on the projects. The PMII must lead to a product that includes significant innovations. It covers research and development work and leads to the development of demonstrators or prototypes, or to the approval of stages with clearly defined specifications and performance. For each stage, the PMII contract outlines the main R&D obligations of the manufacturer. The cooperative framework of the program managed by the manufacturer is also described (consortium, participation of partners and subcontractors, association of public research). PMIIs are financed by the IIA up to a maximum of 50% of the research and development expenditure of consortia members, in the form of grants and/or repayable advances, in compliance with the European framework for R&D aid. Public laboratories participating in PMII are funded by the Agency in a manner appropriate to their status. The first two components of this scheme, developed following the recommendations of Christian Blanc and Jean-Louis Beffa, were then supplemented by a third component aimed at promoting the development of SMEs and mainly innovative SMEs. 2.2.3. Support for innovative SMEs This awareness of the importance of the role of SMEs in the economy is not new; what is more, there is the observation that some SMEs have a major impact on employment because they have strong growth. This is particularly the observation made by Jean-Paul Betbèze and Christian Saint-Étienne in their report Une stratégie PME pour la France, prepared for the Conseil d’Analyse économique in 2006.

Competition and Innovation Policy

79

In this report, they point out the specificity and benefit of this type of SME by using the term “gazelles”, popularized by David L. Birch (2001). Based mainly on a study by Claude Picart (2006) on Les gazelles en France, Betbèze and Saint-Étienne show that any system in favor of SMEs “can only have an effect if it profoundly changes the behavior of the actors, since it is a question both of encouraging entrepreneurial vocations for growth SMEs, but also of ensuring that potential gazelles in France continue their course” (Betbèze and Saint-Étienne 2006, p. 31). But at the same time, they also stress the need to extend the reflection to the whole environment of these companies, insofar as “an economic policy that would only concern itself with increasing the profitability of gazelles without seeing that this also involves thinking about the economic and social fabric surrounding them would, quite quickly, be politically untenable” (ibid.). As a follow-up to this report, various measures to support SMEs have been put in place. Hence, the “SME Growth” plan, presented in December 2006 by Renaud Dutreil, then Minister for SMEs, Trade, Crafts and the Liberal Professions. This plan reflected the idea that SMEs play an essential role in the dynamism of the economy, recalling that around 50,000 SMEs are created each year, but also and above all in terms of employment. At the same time, the life span of these companies is usually relatively short because they fail to develop; hence the idea of targeted public intervention in favor of this type of company. This plan had three components: – creation of a growth company status that is granted to companies that have experienced strong growth two years in a row. This status makes it possible to freeze corporate income tax and gives companies the possibility of deferring the payment of social security contributions due for the last employees hired; – reducing red tape so as to simplify the daily lives of entrepreneurs and not to hinder the growth of companies through unnecessary steps, in particular through the systematic use of new technologies. The objective is to reduce the cost of administrative formalities by 20%; – facilitate financing for SME development. The aim here is to strengthen the markets for venture capital and private equity for SMEs. This last part refers to the various measures put in place to support innovative companies, among which we can mention:

80

Innovation and Industrial Policies

– the strengthening of the Crédit d’impôt recherche (CIR), a research tax credit, which is intended to reduce the financial burden on companies that increase their research effort. Since 2004, the CIR has been composed of a volume share (a percentage of the amount of research expenditure for the year) and an increasing share (a percentage of the increase in research expenditures compared to in previous years). The strengthening of the CIR doubles the volume share, from 5% to 10% of research expenditures. Correspondingly, the share of the increase goes from 45% to 40%; – an incentive to employ young doctoral graduates. In calculating the expenses used as a basis for the CIR, expenses relating to a doctoral candidate’s first year of employment are considered at twice their actual amount. As before, the hiring must be done on permanent contracts and the company’s workforce must not have decreased; – a higher ceiling for patent defense costs. Patent defense costs are included in the calculation of research expenditures. The maximum amount that can be taken into account for this purpose increased to €120,000 for expenditure incurred as of January 1, 2006; – immediate reimbursement of the extended CIR to companies under five years old. New businesses will be able to obtain immediate reimbursement of the CIR for their first five years (instead of the first three up until now); – measures in favor of venture capital. To encourage individual investment in innovation, which involves some risk, two measures are being taken. On the one hand, the reduction of income tax for a subscription to units of Fonds communs de placement dans l’innovation (FCPI). On the other hand, a relaxation of the rules of the Single Personal Risk Investment Company (SUIR) for the benefit of angel investors, in order to allow individuals to develop a business angel activity; – the status of Jeune entreprise innovante (JEI) (Young Innovative Company). To benefit from this status, the company must meet five conditions: be less than eight years old; be an SME in the European sense (< 250 employees, turnover < €40 million or balance sheet < €27 million); be independent (conditions for holding capital); not have been restructured; and incur R&D expenditure representing at least 15% of its expenses. The JEI enjoys a number of important tax and social benefits: an income tax exemption which is total for the first three financial years and then partial (50%) for the following two financial years; an exemption from property tax and/or business tax, upon decision of the local authorities concerned, for

Competition and Innovation Policy

81

seven years; an exemption from taxation on capital gains on the sale of the JEI shares; an exemption from employer social security contributions which applies to salaries paid to employees participating in research (researchers, technicians, R&D project managers, industrial protection lawyers, personnel responsible for pre-competitive tests) and to corporate officers participating mainly in the research project; – research foundations. Considering that in France in 2001 there were only about a thousand foundations whose contribution to the national R&D effort was 0.04% of GDP, while in Great Britain their contribution represented 0.1% of the total research effort and 0.11% in the United States, the government decided to promote the creation of research foundations. Thus, the law of 1 August 2003 on sponsorship, associations and foundations, supplemented by the 2004 Finance Act, created new incentives for companies to invest in research foundations that are recognized as being of public utility. Companies giving to these foundations benefit from a 60% tax credit on donations, up to a limit of 0.5% of turnover. Any excess can be carried forward to the next five years. Approved research foundations are funded by dedicated public funds, up to €1 public for €1 private5. 2.3. The ambiguities of the “new industrial policy” The new industrial policy is presented as a response to doubts expressed against traditional industrial policies. In particular, the creation of competitiveness clusters is part of an analytical approach that believes the main source of innovation lies in a partnership between research and industry, but also between large companies and SMEs. Similarly, the creation of the Industrial Innovation Agency, in charge of developing mobilizing programs, has the function of giving priority to large industrial groups in its promotion of growth based on innovation. However, these programs have a much more cross-cutting content than in the past and focus on the role that private agents must play in their coordination. This new industrial policy has evolved institutionally since its inception, as the government decided to merge the AII with the OSEO group. The objective was twofold: on the one hand, to remedy the weakness in support for innovative medium-sized enterprises and, on the other hand, to set up a one-stop shop within OSEO offering a complete range of assistance adapted 5 For more details, see Besnard et al. (2007) and Jevakhoff and Cavailloles (2017).

82

Innovation and Industrial Policies

to all sizes of innovative enterprises and projects. This merger is undoubtedly a first response to the 2007 Cour des Comptes’ report on Les aides des collectivités territoriales au développement économique (Local and Regional Authorities’ Support for Economic Development), which severely criticizes not only the “fragmented, complex and uncoordinated” nature of business support schemes, but also their “limited economic scope”, which is accompanied by “risks of deadweight and wicket effects”, as well as the “inadequate nature of the evaluation processes” set up. Beyond these criticisms of the real effectiveness of these mechanisms, can we consider that this is a truly new industrial policy? In an attempt to answer this question, it is possible to note, following Jean-Luc Gaffard (2005), that this policy obeys two distinct but complementary logics. The first is a classic logic of agglomeration based on the exploitation of externalities and a selection principle: it is the logic of competitiveness clusters. The second is a development logic involving action on the driving forces of innovation, that is on the articulation between technologies and markets: this is the logic of mobilizing programs for industrial innovation. We will detail these two logics, and then we will come back to the mode of governance that characterizes this new industrial policy. 2.3.1. The logic of agglomeration The logic of agglomeration is specific to each of the poles, it aims to allow the exploitation of the positive externalities that are supposed to result from the mere fact of proximity. Jean-Luc Gaffard shows that, in this case, this logic is “partly similar to the logic of supporting national champions through major technology programs. The aim is to make territories competitive or attractive by favoring companies that are or are likely to set up in them” (Gaffard 2005, p. 4). In this logic of agglomeration or proximity, Gaffard adds that: the policy of competitiveness clusters consists of subsidizing expenditure which would otherwise be lower than its optimal level, as companies are unable to take into account the positive external effects. The role of public authorities is therefore to

Competition and Innovation Policy

83

remedy market failures without intervening in the development of production and market processes (ibid.). This dimension of industrial policy is not fundamentally new, although public discourse has evolved significantly. As Gilles Duranton, Philippe Martin, Thierry Mayer and Florian Mayneris note in their 2008 study, entitled Les pôles de compétitivité, que peut-on en attendre ?: Until the establishment of competitiveness clusters, the aim was to avoid an economic geography that was too concentrated on a few rich regions and to help declining territories. It has moved from a situation where public intervention seemed to assume that economic geography could only be too concentrated and unequal to a situation where specialization, cooperation and clustering of activities in dynamic territories are encouraged by the State. The emphasis on the need to achieve, for a given sector and territory, a certain ‘mass critique’ in order to support international competition, is therefore apparently part of a logic of geographical concentration of economic activities. Similarly, the territorial rapprochement of economic actors is often presented as a necessary condition for improving collaboration and interaction in order to foster innovation (Duranton et al. 2008, p. 13). This dimension of industrial policy is a direct extension of a regional development policy implemented several years ago under the impetus of DATAR. Indeed: before the competitiveness clusters, the policy, although less ambitious, of local productive systems (LPS) had the same type of objective: to strengthen collaboration between companies in a given territory and sector in order to increase their performance. Both are intended to help ‘create new wealth’ and ‘develop employment in the territories’. A quarter of SPLs have been labeled ‘poles’ or have become part of a competitiveness pole (ibid., pp. 12 and 16). However, the study by Duranton et al. points out that this competitiveness cluster policy is a kind of reversal of the previous spatial

84

Innovation and Industrial Policies

planning policy. The latter sought to avoid an economic geography too concentrated on a few rich regions and to help declining territories, while the new policy encourages the concentration of activities on the most dynamic territories. This moves us from an objective of equity to an objective of efficiency. On this new objective of efficiency, the study is very critical since it underlines that: Cluster policies can also have negative effects when the political economy of decisions is taken into account, sometimes guided by political rather than economic concerns, or when the risk of excessive specialization in regions that become dependent on the fate of a sector is taken into account, particularly when workers are not very mobile, as is the case in France (ibid., pp. 20–21). More generally, these authors show that the expected efficiency gains are relatively modest since doubling the level of specialization in a given activity and territory would result in an increase of about 5% in the productivity of the firms concerned. From this they deduce that “public policies that try to affect the location choices of companies are not very effective. The gains to be expected from a cluster policy are therefore relatively modest: there is no economic miracle to be expected from such a policy” (ibid., pp. 21–22). While the quantified results of their analysis do not allow us to conclude that “competitiveness cluster policy will have no effect on the attractiveness of territories or on the productivity of companies”, they nevertheless lead us to “temper the enthusiasm that this perspective has generated among politicians on both the left and the right” (ibid., p. 23). 2.3.2. The logic of development Unlike the agglomeration logic, this development logic is mainly based on the PMII. These programs are based a priori on two complementary principles: on the one hand, a principle of concentration and, on the other, a principle of variety selection. However, it is above all the principle of concentration that dominates since it is basically a question of gathering resources around a small number

Competition and Innovation Policy

85

of actors to ensure the development of industrial projects, which are often well defined around new products or new technologies. From this point of view, these PMII therefore appear “to be in line with the traditional policy of major programs, even if they are no longer part of the triptych of public research/public company/public procurement” (Gaffard 2005, p. 5). On the contrary, they stress the need to identify high-growth markets, and stress the essential role of company networks, but “they remain mainly dedicated to promoting oligopolistic concentration, in this case at European level, without really taking into account the necessary variability of industrial structures” (ibid.). Indeed, the underlying industrial strategy does not fundamentally aim to reduce or share R&D costs since “the main challenge of the cooperation set up is to help build markets whose existence justifies R&D commitments” (ibid.). These oligopolistic concentrations are based on the idea that their component companies remain in competition and therefore have no incentive to engage in vertical integration. This type of partnership industrial structure obviously raises the problem of the role and future of SMEs in relation to large companies. Moreover, since mergers formed under this type of strategy must take place in a context of global competition, they inevitably raise the question of their size and scope, but also the question of their sustainability. Indeed, the challenge is how to reconcile the development of innovations and the increasing returns they generate with the maintenance of competition within these industrial structures, mobilizing programs being the main instrument dedicated to supporting the growth of SMEs by enabling them to capture significant market shares. In total, the two logics of agglomeration and development are therefore nothing really new compared to traditional industrial policy practices. The only new element is the governance mode put in place, which is supposed to ensure the development of interactions between partners and establish the conditions for the selection of projects proposed for public funding. 2.3.3. A new mode of governance In the field of governance, the new industrial policy has the particularity of not being part of the traditional interventionist approach, nor of an approach that would be limited to the definition of an appropriate regulatory framework. It consists of delegating to companies the task of defining the

86

Innovation and Industrial Policies

poles to be labeled or the projects to be financed and developing the appropriate organizational mode. The State, through the agencies and the Fonds unique interministériel (FUI), limits itself to choosing among the various projects proposed and financing those that it deems the best. The FUI is therefore intended to support collaborative R&D projects aimed at developing new products or services that could be launched on the market in the short or medium term. It should also be noted that the FUI is now managed by the Banque publique d’investissement (BPI France), created in 2012 by the merger of OSEO and CDC Enterprises, which brings together the private equity activities of Caisse des dépôts, and since 2013, the Fonds stratégique d’investissement (a sovereign investment fund created in 2008). The State also retains responsibility for evaluation and monitoring of the various projects. However, the success of such an approach requires at least two conditions to be met. The first is that “the governance structure of the poles must effectively promote the emergence of variety, in particular by closely associating public and private research, large and small companies all more or less directly involved in R&D activity, without hindering selection or entry and exit mechanisms” (Gaffard 2005, p. 7). The second condition is that “project selection and evaluation procedures must lead to the definition of success and failure indicators likely to allow the necessary interruptions or reorientations” (ibid.). Clearly, these conditions are not yet fully met. Thus, Jean-Luc Gaffard notes that “the choice of poles has partly obeyed the old criteria of industrial policy, partly the new ones: this is probably the meaning of the distinction made between global poles or poles with a global vocation and the other labeled poles. The governance structures implemented, while formally similar, reflect very different degrees of progress” (ibid.). It also adds that “the division of tasks and competences between the central and decentralized levels of government is not really clarified [...]. The problem is not that the regions take the place of the State, but that each one plays its role correctly. It is undoubtedly up to the regions to participate in financing, but without having to bear the cost of anything that would not be financed by the State” (ibid.). More generally: “Shared competences appear to be an essential issue for the necessary institutional clarification. Clarification is far from being achieved by referring to the notion of subsidiarity” (ibid.). It may be recalled that the principle of subsidiarity reflects the idea that the responsibility for public action, when necessary, should be allocated to the smallest entity capable of solving the

Competition and Innovation Policy

87

problem on its own. It is therefore a concern to ensure that no higher level is used to achieve what can be achieved equally effectively on a smaller scale, that is to constantly seek the relevant level of public action. The same type of remarks can be addressed to programs mobilizing industrial innovation. The novelty of these programs must also lie in their mode of governance. However, it does not seem that there is a significant difference in this area from previous programs. More specifically, it does not seem that the governance conditions adopted at the AII level make it possible to avoid “the risk of capture of public intervention by large groups” (ibid.). Nevertheless, this new industrial policy was complemented by the decision to proceed in 2009 with a “big loan” that led to the implementation of what is still today called the Programme d’investissement d’avenir (PIA). 2.4. The Programme d’investissement d’avenir (PIA) In June 2009, the decision in principle was taken to raise a national loan to finance strategic investments designed to prepare France for the 21st Century. In August 2009, a commission chaired by Alain Juppé and Michel Rocard was set up. It was mandated by the President of the Republic to identify and assess future investment needs. On November 19, 2009, the Commission submitted its report, which defined seven strategic priorities and proposed a new form of governance for public action. This report is entitled: Investir pour l’avenir. Priorités stratégiques d’investissement et emprunt national. In December 2009, the President of the Republic decided to launch a €35 billion major loan to finance five strategic priorities: higher education and training, research, industry and SMEs, sustainable development and digital technology. Management of this program was entrusted to a body created for this purpose in January 2010: the Commissariat général à l’investissement, which was renamed the Secrétariat général pour l’investissement (SGPI) in 2017. The latter reports directly to the Prime Minister’s Office and his Commissioner General (or now his Secretary General) is appointed to the Council of Ministers. It is therefore from January 2010 that the major loan has been implemented under the name of the Programme d’investissement

88

Innovation and Industrial Policies

d’avenir (PIA). We will first present the main lines of this program, recalling the logic of the Juppé-Rocard report (section 2.4.1), then we will examine its implementation modalities (section 2.4.2) and end with an analysis of the problems raised by the articulation between the PIA and the new industrial policy previously put in place (section 2.4.3). 2.4.1. The logic of the Juppé-Rocard report To understand the logic of the PIA, it is first necessary to recall the logic of the Juppé-Rocard report, since it is on this report that this program is based. This report develops three main ideas. The first is based on a twofold observation. On the one hand, to meet the challenges of tomorrow, France must invest. Indeed, in a context of heightened international competition, the lack of investment today increases the risk of a long-term dropout. On the other hand, with the deterioration in public finances, the share of investment in public expenditure has declined. In the early 1990s, this share accounted for about 6% of public expenditure, but in 2009 it was only 5%. The increase in investment by local authorities over the past decade or so has not been able to offset this steady decline in government investment. Faced with this observation, the Juppé-Rocard report considers that “if we do not resign ourselves to this decline in investment, if we believe on the contrary that we must invest for the future and that it is urgent to do so, then borrowing is essential. Borrowing is indeed justified when it comes to investing with a view to a future return. And borrowing makes it possible to act quickly”. The second idea that emerges from this report is the stated desire to build a new model of more sustainable development: “If we do not want to resolve to the collapse of growth and employment, we must find new drivers, new sources of development.” The first requirement defended by the Juppé-Rocard report is therefore to focus on higher education and scientific and technological research. However, relying on research is not enough, it is also necessary to know how to value its results and develop research oriented towards industrial development. Similarly, investing in knowledge is not enough in itself, it is also necessary to be able to meet the ecological challenge. All in all, it is a question of moving away from old development patterns and moving towards a new model of sustainable development based on both grey matter and the “green” economy. The authors of this report therefore need to take a long-term perspective in order to prepare for the future through a targeted and exceptional investment effort.

Competition and Innovation Policy

89

To achieve this objective, the Commission’s projects were subject to the following criteria: – invest in areas that represent a medium- or long-term strategic challenge with a ‘transforming’ logical, with a view to the transition to a new development model; – focus on areas where France has comparative advantages; – finance exceptional investment expenditure likely to make a difference; whenever possible, there should be a new asset in return for the State’s contribution; – to compensate for market failures (financing problems linked to an overly long horizon or a high risk, etc.) by joint intervention or in support of private action; – act on sectors where government investment can unlock or reach critical size and lead to positive externalities; – seek, whenever possible, a leverage effect with regard to private financing as well as other public, local or Community financing; – ensure a return on investment, financial or socio-economic, and be able to be the subject of an evaluation of public intervention (Juppé-Rocard 2009, pp. 23–24). In this perspective, the Juppé-Rocard report defines a number of strategic priorities aimed at ensuring the transition to a new development model and corresponding to a €35 billion government investment. Axis 1 Axis 2 Axis 3 Axis 4 Axis 5 Axis 6 Axis 7

Support higher education, research and innovation Fostering the development of innovative SMEs Accelerate the development of life sciences Develop low-carbon energies and efficiency in resource management Bringing out the city of tomorrow Inventing the mobility of the future Investing in the digital society Total

€16 billion €2 billion €2 billion €3.5 billion €4.5 billion €3 billion €4 billion €35 billion

Table 2.1. Strategic investment priorities for national borrowing (source: Juppé-Rocard 2009, p. 14)

90

Innovation and Industrial Policies

The report also states that “by leveraging private, local and European financing, the national loan should ultimately correspond to a total investment of more than €60 billion” (ibid., p. 15). Investments proposed as part of partnership projects between public operators and the private sector and interventions in the form of loans should have a leverage effect of between one and three. Thus: the investment expenses included in the report give rise to the creation of nearly 60% of assets. Other expenses are accompanied by a return requirement. In all cases, the expenditure chosen is profitable either directly (dividends, royalties, interest, etc.) or indirectly (tax revenue generated by increased economic activity) for the State and provides socioeconomic benefits for the community (ibid.). The third fundamental idea of the Juppé-Rocard report is to propose the implementation of a rigorous governance system. On the financial side, he suggests that: the funds raised by the national loan are allocated to managing bodies and managed in a way that is watertight in relation to the rest of the budget. They must have an additional effect over and above the usual budgetary funding and not replace funding allocations to the managing or beneficiary bodies of the bodies financed. They cannot therefore be used to finance the salaries of civil servants (ibid.). Similarly, at institutional level, the report recommends “the establishment, under the Prime Minister, of a National Loan Supervisory Committee composed of parliamentarians, qualified personalities and representatives of the ministries concerned” (ibid.). In addition, the report specifies that the implementation of the investment plan must be the subject of a contract between the State and the managing bodies. In this contractual framework, it proposes to set up a National Loan Supervisory Committee, “composed of parliamentarians, qualified personalities and representatives of the ministries concerned” (ibid.), which would be responsible for monitoring the management of the funds and steering the evaluation of the actions financed. This Committee should also

Competition and Innovation Policy

91

report periodically to Parliament on the use of funds and publish the results of its evaluations. It is these three ideas and the main proposals of the Juppé-Rocard report that have been included in the Programme d’investissement d’avenir (PIA), with however some modifications that are not negligible. 2.4.2. The PIA architecture The logic behind the PIA architecture is based on three pillars: the objectives pursued, the methods of intervention and the mode of governance. With regard to the objectives pursued, the approach that characterizes the PIA is the same as that of the Juppé-Rocard report. As indicated in the Rapport relatif à la mise en œuvre et au suivi des investissements d’avenir, annexed to the 2011 Finance Bill: The purpose of this program is to prepare France for the challenges of tomorrow, by investing €35 billion in higher education and training, research, industrial sectors and SMEs, sustainable development and digital technology. In total, with the leverage effect of other financing, and in particular private co-financing, the expected investment program is in the order of €60 to 65 billion. The only real difference is that instead of selecting seven priority axes, the program targets only five: – higher education and training (€11 billion); – research (€7.9 billion); – industrial sectors and SMEs (€6.5 billion); – sustainable development (€5.1 billion); – digital economy (€4.5 billion). These five strategic priorities correspond to a transversal approach to investment in education and research, but they also cover more thematic dimensions where innovation must enable the French industrial fabric to ensure a competitive positioning in the long term (digital economy, clean

92

Innovation and Industrial Policies

energies and less polluting transport, health and biotechnology). Such targeting also reflects the desire to strengthen links between the various innovation actors (in particular between public and private research and between SMEs and large groups), in order to create genuine “ecosystems”. Finally, it is worth noting the importance given to teaching and research, which accounts for more than half of the investment effort (54%). As far as the modalities of public intervention are concerned, the PIA marks a real qualitative turning point in the design of industrial policy. First of all, it is based on a logic of continuous learning. There is no pre-established timetable but a progressive commitment of resources according to the nature and intermediate evaluations of the projects. Another new element is that the approach is explicitly based on a bottom-up and selective strategy, which is based on national calls for projects. The selection process focuses mainly on the intrinsic quality of projects, but also takes into account the existence of ripple effects. Thus, the financing modalities adopted aim to encourage the creation of assets through a co-investment approach, that is subsidies represent only 40% of the financing and are accompanied by financial return or incentive clauses for the success of the subsidized technology. These modalities therefore aim to promote leverage effects towards private financing as well as financing from local authorities, Europe and other public actors, so that overall investment should be between €60 billion and €70 billion, for €35 billion in public credits. However, the real originality of the PIA lies above all in the governance method adopted. Indeed, this mode of governance is partly different from that provided for in the Juppé-Rocard report since it is based on the creation of a new institution: the Commissariat général à l’investissement (CGI), which is placed directly under the authority of the Prime Minister. The CGI has two main missions. In general, it ensures the consistency of the State’s investment policies. More specifically, it is responsible for managing the PIA and in particular for: – preparing government decisions on agreements between the State and operators responsible for the management of funds and the execution of investments; – coordinating the preparation of specifications accompanying calls for projects and check their consistency with government action in the field of investment and public policy reform;

Competition and Innovation Policy

93

– coordinating the appraisal of investment projects; – formulating opinions and proposals for the arbitration of the President of the Republic and the Prime Minister; – ensuring that investments, and in particular their profitability, are evaluated a priori and a posteriori; – drawing up an annual review of the implementation of investment programs. The work of the CGI is overseen by a Supervisory Committee, composed of eight parliamentarians and eight qualified personalities. This Committee was initially chaired by Alain Juppé and Michel Rocard. The governance of the PIA is characterized by a double contractualization. The first between the CGI and the various operators responsible for calls for tenders and the selection of projects to be financed, the second between operators and project leaders. These operators are the various public institutions in charge of the topics covered by the PIA, such as OSEO, ANR (Agence nationale de la recherche), ADEME (Agence de l’environnement et de la maîtrise de l’énergie), ANDRA (Agence nationale pour la gestion des déchets radioactifs), CEA2 (Commissariat à l’énergie atomique et aux énergies alternatives), ANRU (Agence nationale pour la rénovation urbaine), CDC (Caisse des dépôts et consignations), CNES (Centre nationale d’études spatiales), ONERA (Office national d’études et de recherches aérospatiales), ANAH (Agence nationale de l’habitat). The 35 projects initially selected were the subject of agreements signed between the General Investment Commissioner and the 10 public operators in charge of the selection process. These agreements specified, action by action, the objectives pursued, the evaluation criteria, the selection processes, then the monitoring and finally the ex post evaluation. It is then possible to observe that the Office of the Commissioner General for Investment has in fact played a real role as an integrator. 2.4.3. The link between the PIA and the competitiveness clusters As shown by the ex post evaluation of the PIA, led by Jean-Louis Levet and Claude Mathieu in 2013, this program can be considered as a new step in the evolution of the concept of industrial policy (Ravix 2013). Although it is clear that the PIA cannot replace the new competitiveness cluster policy,

94

Innovation and Industrial Policies

there is no doubt that it will strengthen and complement it in at least three directions. The first one confirms and even accentuates the choice in favor of a bottom-up approach. The second reinforces the idea of a shift from industrial policy to innovation policy by affirming the need for a sustained effort (more than 50% of funding) in favor of higher education and research. The third direction is to maintain the idea of the need for greater support for industry and SMEs, which also requires a more explicit display of certain sectoral priorities (biotechnology, sustainable development, transport, digital, etc.). But at the same time, the PIA stands out clearly from the competitiveness cluster policy in at least three areas. The first is that it introduces an important change in the institutional environment with the creation of the CGI, since it is this institution that is now responsible for ensuring the coherence of the State’s various investment policies. The second area concerns the desire expressed in the Juppé-Rocard report to set up “exemplary governance”. This new governance is characterized by an accountability approach, based on a dual contractualization between, on the one hand, the CGI and the operators (ANR, ADEM, CDC, etc.) and, on the other hand, between the operators and the project leaders; it is itself supplemented by a monitoring of the selected projects, a Monitoring Committee and a procedure for evaluating the system. The third area in which the PIA differs strongly from the competitiveness cluster policy is an explicit and voluntary absence of reference to a spatial planning logic, which was one of the traditional components of industrial policy, but also to a cluster logic specific to competitiveness clusters. Indeed, when Jean-Luc Tavernier, Deputy Commissioner General, mentioned the links between future investments and competitiveness clusters in 2011, he specified that while “great attention is paid to the implementation of the PIA in the territories”, on the other hand, there is “no logic of spatial planning for all that, but of optimizing the potential of each individual, through the filter of excellence” (Tavernier 2011, p. 9). By explicitly excluding any territorial dimension of the PIA, but by recalling that “competitiveness clusters are recognized as a place to focus and stimulate excellence and as an important indicator of the quality of territorial dynamics” (ibid.), Jean-Luc Tavernier confirms the fact that no specific role is assigned to competitiveness clusters, even if they are strongly encouraged to actively participate in the various priorities of the PIA (shared innovation

Competition and Innovation Policy

95

platforms, IRT, sectoral measures, etc.). Competitiveness clusters are thus losing the driving role they had previously to become mere facilitators or partners in projects that structure territorial dynamics. 2.5. Conclusion The new mode of public intervention, which has gradually developed with the establishment of the competitiveness clusters and the PIA, marks an important step in the evolution of the concepts of industrial and innovation policy, for two main reasons. First, it reinforces the “matrix” conception, initiated earlier, by more explicitly affirming the need to combine a vertical dimension of targeted industrial policy with a horizontal dimension of innovation policy. Second, it reflects the choice in favor of a more proactive policy, which is expressed not only in the importance of the financing mobilized and the expected leverage effects, but also in the implementation of a “new governance”, placed under the responsibility of the CGI. The latter is thus given a central role in the management of this new industrial policy. The fundamental question raised by this new “matrix” concept obviously concerns the coherence of the overall system, that is the way in which the industrial development strategy and the research and innovation strategy are actually articulated, but also its relevance in combating unemployment and, above all, deindustrialization, which is apparently the main cause.

3 Reindustrialization Through Innovation

As the Commission nationale d’évaluation des politiques d’innovation (National Commision for the Evaluation of Innovation Policies) (CNEPI 2016, p. 9) notes, the consensus around the predominant role of innovation in the process of economic growth has resulted in a major institutional reorganization with the establishment of two new actors. The first is the Secrétariat général pour l’inverstissement (SGPI) (General Secretariat for Investment), whose function is in particular to manage the Programme d’investissement d’avenir (PIA). The second is the Banque publique d’investissement (Bpifrance), a public investment bank, which is in charge of supporting and financing companies’ innovation efforts. This observation could suggest that the principle of an innovation policy has fully replaced that of an industrial policy. However, such an interpretation cannot be accepted since, during the same period, the marked action in favor of innovation has gradually been accompanied by a renewed interest in questions relating to the organization, structuring and development of the industry. This renewal was only the direct consequence of the realization of an inexorable movement, if not decline, at least of “industrial disengagement” (Cohen and Buigues 2014). The alarming nature of these consequences has led to the mobilization of all socio-economic partners, which led to the organization, at the end of 2009, of the États généraux de l’industrie (General Committee for Industry). At the end of these États généraux, the bitter conclusion was that industry had lost the status of a major national player which it had succeeded in conquering during the post-war period:

Innovation and Industrial Policies, First Edition. Joël-Thomas Ravix and Marc Deschamps. © ISTE Ltd 2019. Published by ISTE Ltd and John Wiley & Sons, Inc.

98

Innovation and Industrial Policies

French industry has certainly continued to adapt to the changes in the economic world resulting from globalization, with remarkable success in certain sectors of activity, but its overall capacity has been reduced as a result, while priority has been given in France, as in other developed countries, to the service and finance sectors, considered to bring more value added per agent, to be safer and more profitable, and as the massive opening of world trade, encouraged by standardization and lower transport costs, has enabled certain emerging countries to become industrial champions (États généraux de l’industrie 2010, p. 9). By mobilizing all socio-economic partners, the États généraux de l’industrie succeeded in supporting the need to “give the country back a real industrial ambition” (ibid., p. 63) and, to this end, to direct public authorities towards a real renewal of industrial policy in France: The idea of industrial policy, long criticized as archaism, is once again becoming a respected reference. An objective analysis of the situation in countries as different as China or the United States shows that the States still play a key role when it comes to developing a global industrial power such as China or in ensuring the development of new global industrial champions, as in the software industry in the United States (ibid., p. 64). In this perspective, the Etats généraux report recommends two new measures that are different but complementary. The first is institutional. The aim is “to set up a Conférence nationale de l’industrie (National Conference on Industry), bringing together, around the State in an inter-ministerial framework, all the actors concerned by the future of industry, in particular the social partners, professional organizations, national and regional public actors, specialized bodies, experts and creators” (ibid., p. 68). This new institution, which is responsible for monitoring the actions implemented, but also for serving as a framework for reflection and action for industry players, is not a substitute for competitiveness clusters. On the contrary, the report considers that, “given their structuring role in the industrial sector, competitiveness clusters should, in particular, be involved in the work of the National Conference” (ibid.).

Reindustrialization Through Innovation

99

The second measure is more original since it is in line with the old notion of the sector. Indeed, the concept of the industrial sector was mainly developed in the 1970s to refer to “all the stages of the production process that lead from raw materials to satisfying the end need of the consumer, whether this end need is for a material good (such as food, housing, etc.) or a service (such as transport, communications, health, leisure)” (Stoffaës 1980, pp. 86–87). The report by the États généraux thus notes that it is “at the level of sectors that the reflections on market developments and their conditions of development can most usefully be conducted to feed a coherent national industrial strategy” (ibid., p. 69). This measure consists of “encouraging all major branches or industrial sectors to rethink their governance on the basis of a new mapping to be drawn up by the market sector (aeronautics, railways, automobiles, nuclear energy, health, information and communication technologies, etc.) combined with the branch approach more oriented towards careers, technologies (mechanics, plastics, electricity and electronics, etc.) or services (engineering, IT, logistics, etc.)” (ibid., p. 86). These two measures, which were the result of a broad reflection carried out at the time by all economic and social actors, constituted the foundations for a major change in the way public authorities conceive the modalities of action in the fields of innovation and industrial production. Following the recommendations of the États généraux de l’industrie, the Conférence nationale de l’industrie (CNI) was created in June 2010. It is an advisory body, responsible for informing and advising the government on the situation of the industry. To fulfill this function, the CNI has been structured into twelve strategic sector committees and four transversal working groups, which are tripartite bodies bringing together representatives of companies, employees and the administration. The twelve strategic committees concern: the aeronautics sector, the automotive sector, the consumer goods sector, the eco-industries sector, the naval sector, the chemicals and materials sector, the rail sector, the agri-food industry sector, the health industry sector, the fashion and luxury sector, the nuclear sector and the ICST sector. The four groups are: review and foresight, regulation and simplification, R&D and innovation and jobs and skills. This chapter is devoted to an analysis of the implications of this new ambition, which aims to articulate innovation policy and industrial policy. The first section will focus on the study of its institutional evolution and the problems raised by the return of the concept sector (section 3.1). The second

100

Innovation and Industrial Policies

section will address the question of the nature of industry, which is at the heart of the theoretical problems posed by any attempt to articulate a horizontal policy to support innovation and a vertical policy in favor of industry (section 3.2). Finally, the third section will specify the main theoretical arguments justifying such a renewal of State intervention in favor of industry (section 3.3). 3.1. The affirmation of an industrial ambition The CNI published its first annual report on 2011 in 2012. It was entitled “Together, reindustrializing France for growth and employment”. After noting that the deterioration of French industry had continued in 2011, it identified “nine challenges and proposals for the sustainable growth of French industry” (CNI 2011, p. 11). The report thus showed “contrasting and cyclical” socio-economic results for industry and the various sectors. As it also pointed out: All the twelve strategic sector committees set up to lead the discussions on the sectors that were identified as priorities have noted the need to work more closely between sectors, to confront the many challenges facing the sectors and to take advantage of the opportunities that are opening up in terms of new markets (ibid., p. 39). The organization of the sectors was therefore presented as a means not only of gaining in competitiveness, but also of establishing in a sustainable and concerted manner “win-win relations between principals and subcontractors” (ibid.). This new approach to the industry reflects a significant change in perspective that is expressed in two different but complementary ways. It is indeed possible to show, first, that the challenge of reindustrialization now leads to industry, and no longer the market, as the fundamental concern of economic policy (section 3.1.1) and, second, that such a change of orientation is considerably reinforced by the use of the concept of the industrial sector, which places the question of the organization of productive structures at the forefront (section 3.1.2).

Reindustrialization Through Innovation

101

3.1.1. A new concern: industry The change of government that occurred after the 2012 presidential election did not fundamentally change the nature of the diagnosis or the measures recommended. On the contrary, it has strengthened it politically, with the creation of a Ministry of Productive Recovery and, technically, with the definition of “a genuine productive pact for competitiveness, growth and employment”, the elaboration of which has been entrusted to Louis Gallois, Commissioner General for Investment (Gallois 2012, appendix). In his report of November 5, 2012, Louis Gallois reiterated several of the recommendations of the 2011 Annual Report by the CNI, including the need to create a “competitiveness shock”, the only one capable of restoring confidence and boosting investment momentum (Gallois 2012, p. 23). It is with this in mind that he recommends in particular strengthening the resources of the CNI so that it can “fully play its strategic role, particularly with regard to sectors” (ibid., p. 21). On 6 November, the measure was ratified since Decision No. 11 of the Pacte national pour la croissance, la compétitivité et l’emploi (National Pact for Growth, Competitiveness and Employment) proposes to “overhaul the Conférence nationale de l’industrie (CNI) to implement pacts between companies in the same sector” (Premier ministre 2012, p. 3). This decision has the effect of changing its name since the press release of the Ministry of Productive Recovery, dated February 5, 2013, states that, “in accordance with the decisions of the National Pact for Growth, Competitiveness and Employment, the Conférence nationale de l’industrie becomes the Conseil national de l’industrie (National Council for Industry) with the objective of promoting pacts between industries that depend on each other”. The same press release states that: As a true industry parliament, the CNI brings together industrialists and trade union organizations around the State under the chairmanship of the Prime Minister. It structures its work by sector to identify specific concrete actions that strengthen each industry in the global competition. More generally, by adopting the National Pact for Growth, Competitiveness and Employment, the public authorities defined “eight levers of competitiveness” which brought together 35 concrete decisions. Among them, two main measures were directly aimed at changing the tax

102

Innovation and Industrial Policies

and financial environment for companies. The first was the introduction of a “Competitiveness and Employment Tax Credit, the Crédit d’impôt pour la compétitivité et l’emploi” (CICE) (ibid., p. 1). By this measure, the government intended “to give companies the means to reposition themselves in a sustainable offensive repositioning in international competition, for those directly exposed to it, and for all companies, to contribute to the competitiveness of the national economy by moderating their costs” (ibid.). The second measure was based on the observation that “too many very small businesses (VSBs), small and medium-sized enterprises (SMEs) and medium-sized businesses (MSBs) are facing financing difficulties, which hinder their ability to innovate, export, develop or even threaten their survival” (ibid.) With this in mind, the government had to decide to create, at the beginning of 2013, the Banque publique d’investissement (Public Investment Bank) (BPI) “to offer companies, with priority for VSBs, SMEs and MSBs, a proximity financing service using a wider range of financial instruments and advice to intervene at all stages of the company’s development” (ibid., p. 2). This new mechanism, composed of the Conseil national de l’industrie and the National Pact for Growth, Competitiveness and Employment, thus perfectly expresses the return of industry and enterprises to the heart of the public authorities’ concerns. This return is all the more marked as it was first completed, in 2013, by the launch of the Nouvelle France industrielle (NFI) to support the move upmarket of companies and help them position themselves in the markets of the future, by defining “34 industrial recovery plans”. The objective was to enable France to “become a pioneer in the third industrial revolution, at the crossroads of ecological and energy transitions on the one hand, and numeric and digital on the other” (Gouvernement 2014, p. 3). Then, in April 2013, it was also strengthened by a new institution: the Commissariat général à la stratégie et à la prospective (General Commissariat for Strategy and Foresight), more commonly known as France Stratégie, which replaced the Conseil de l’emploi, des revenus et de la cohésion sociale (CEREC) and the Centre d’analyse stratégique, which had itself replaced the Commissariat général au Plan in 2006. The function of France Stratégie is to contribute to “the determination of the main orientations for the future of the nation and the medium and long-term objectives of its economic, social, cultural and environmental development, as well as to the preparation of reforms” (Communication of the Council of Ministers of April 17, 2013). Finally, while the whole system reflects the affirmation of a real desire to reverse the deindustrialization movement

Reindustrialization Through Innovation

103

through the adoption of an ambitious industrial policy, innovation policy is not being ignored. In the same year, the “Innovation 2030 Commission” was set up, whose mission is to encourage and support innovation projects that meet society’s essential needs by 2030 (Lauvergeon 2013, p. 47). This raises the idea of the need for complementarity between the vertical nature of industrial policy and the horizontal nature of innovation policy. This concept is explicitly mentioned in the government plan entitled Une nouvelle donne pour l’innovation (A New Deal for Innovation), presented in Bercy on November 5, 2013 by the Minister of Productive Recovery, the Minister of SMEs and Digital Technology and the Minister of Higher Education and Research. It states that innovation is a key factor in the economy’s move upmarket, but that “despite its strengths, France risks falling behind in the global competition for innovation” (Gouvernement 2013, p. 3). This Nouvelle donne pour l’innovation proposes 40 measures that are presented as complementary and reinforcing the sectoral approach of the NFI and the seven ambitions of the “Innovation 2030” Commission. As the latter states in its report: The Commission’s exercise is complementary to the Nouvelle France industrielle project, which implements 34 plans defining growth drivers for industrial sectors in today’s markets. The Commission, for its part, wants to encourage, within ten years, French industrial leaders on an international scale, in specific sectors, by concentrating resources on key areas (Lauvergeon 2013, p. 6). Added to the 35 measures of the National Pact for Growth, Competitiveness and Employment, the whole system seems all the more redundant as some of these measures stem from the 44 recommendations suggested by the Assises de l’entrepreneuriat, which, between January and April 2013, brought together around the government a group of entrepreneurs, trade unions and employers’ organizations; while others are directly inspired by the conclusions of the report by Jean-Luc Beylat and Pierre Tambourin (2013), entitled: L’innovation, un enjeu majeur pour la France (Innovation, a Major Challenge for France). While the idea of complementarity between industrial policy and innovation policy is beginning to emerge, it is still largely confused, undoubtedly because of the multiplication of reports, plan announcements,

104

Innovation and Industrial Policies

measures and recommendations that are often redundant; but above all because of the ambiguities attached to the return, in force, in the discourse of public authorities and industry players, of the old concept of the sector. 3.1.2. The return of the sector concept The sector is the key concept of the mechanism set up following the États généraux de l’industrie which led to the creation of the Conférence nationale de l’industrie in 2010, and one of its main objectives was to “strengthen the governance of industrial policy, particularly by sector, by creating a structure involving all stakeholders” (CNI 2011, p. 5). It is within this consultative body, which became the Conseil national de l’industrie in February 2013, that the main orientations of industrial policy are now being developed through the fourteen strategic sector committees. It should be noted that there were twelve such strategic industry committees in the first report of the 2011 Conférence nationale de l’industrie, and that they increased to fourteen as of the 2013 report of the Conseil national de l’industrie. In doing this, the public authorities are bringing an old idea, already mobilized as an industrial policy tool, back to the forefront. As Thibaut Bidet-Mayer and Louisa Toubal perfectly summarize it: It was from the 1960s that this term spread with the formulation of post-war policies. The sector was then used to describe the different operations required to move from a raw material to a finished product and to compensate for the inadequacy of approaches in terms of sectors or branches. In the 1980s and 1990s, the globalization of the economy called into question the relevance of the concept of the sector envisaged in a dimension that was too exclusively vertical (Bidet-Mayer and Toubal 2013, p. 21). This notion of sector thus has two characteristic features. The first is that it has undeniable empirical relevance since it has been adopted, not only by public authorities, but above all by industrialists, as the appropriate framework for properly addressing all issues relating to industry. However, although it seems to be perfectly suited to the analysis of industrial phenomena, it remains a multifaceted concept and therefore difficult to define. This difficulty is related to the fact that, from the very beginning of its employment, this term was used for very different purposes.

Reindustrialization Through Innovation

105

It should be noted, however, that none of the CNI’s various annual reports propose a precise definition of the concept of the sector, without this raising any problems. The various sectors are identified in a purely empirical way, because “the imprecision of this concept makes it possible to use it with flexibility” (Gallois 2013, p. 11). The approach is therefore essentially pragmatic. The second characteristic feature of the concept of the sector is that it is a French specificity, which does not seem to have an equivalent in English or German. Some authors thus extend the specificity of the concept to the specificity of French industrial policy. This is confirmed by this remark: France oscillates between the American dream of Silicon Valley, where disruptive innovations are driven by start-ups, the German dream of a well-established industrial Mittelstand that performs well in incremental innovation, and a French tradition of industrial planning in regal sectors (Beylat and Tambourin 2013, p. 6). They can then deduce that such specificity generates ambiguities insofar as “this oscillation blurs France’s representation of innovation because it mixes breakthrough innovation, incremental innovation and ‘strategic industrial policy’” (ibid.). Two perspectives are available a priori to avoid these ambiguities. The first one consists of purely and simply rejecting the concept of the sector. This is the position adopted by David Encaoua, who considers that the notion of sector favors a vertical division in which the emphasis is placed on the relations between the different stages of a production process and that these relations can be assimilated to those between a principal and a subcontractor. However, for the author, these relationships have two limitations: On the one hand, they are not stable over time insofar as the technologies themselves are subject to significant variations, particularly due to innovation. On the other hand, these relationships are characterized by market imperfections, due to opportunistic behaviors that give rise to hold-up phenomena (Encaoua 2013, p. 52).

106

Innovation and Industrial Policies

In this way, it manages to dismiss the relevance of the concept of the sector by stressing its unstable and non-competitive nature, preferring an approach in terms of the market and therefore competition policy since: In a market economy where the various entities appearing in the upstream and downstream stages of production are necessarily autonomous, the State has few means to establish harmonious relations between them. It can only act by introducing new trade rules (ibid., p. 54). The same attitude can be found in Jean-Luc Gaffard, but for radically different reasons, since he proposes “to challenge the relevance of the notion of sector to replace it with that of a cluster, which seems to correspond better to the need for industrial policy to recognize the pre-eminent role of the company in defining strategic choices” (Gaffard 2013, p. 84). The argument is somewhat symmetrical to the previous one since, on the one hand, it retains the idea that a sector can only be defined by the stability of the technical relationships that define it and, on the other hand, it mobilizes the theory of transaction costs to show that “companies’ strategic choices are far from being dominated by technology, which is also a fixed technique. The structure of the industrial fabric is constantly evolving as a result of these choices and the constraints that determine them. In other words, sectors are more a result of innovation processes than technical frameworks that drive strategic choices” (ibid., p. 86). Thus, rather than the notion of sector, the author prefers the notion of cluster, which is for him “a tool that aims to develop voluntary cooperation between companies and which constitutes a network of skills” (ibid., p. 88). The interest of the cluster lies in the fact that it is the companies that determine its configuration and it results that “the creation of skills resulting from the network organization promotes the capillarity and the progressive entry of its various members into new fields of activity” (ibid.). This last point of view is in line with that adopted in the BeylatTambourin report, which prefers to retain the notion of ecosystem because it seems to be the only one that offers the possibility of simultaneously capturing the horizontal dimension of the cluster and the vertical dimension of the sector. As this report does indicate: Innovation is particularly stimulated within local ecosystems, generally at the scale of a metropolis, anchored in a territory.

Reindustrialization Through Innovation

107

These local ecosystems are part of the strategy of the region to which they belong. At the same time, their economic sectors position them within sectors on a supra-regional scale, with a structuring role for competitiveness clusters. A policy in favor of innovation ecosystems must take into account the articulation between these three scales (Beylat and Tambourin 2013, p. 24). By using this notion of innovation ecosystems, the authors of this report thus succeed in showing that the implementation of transversal measures, characteristic of horizontal innovation policies, does not necessarily make it possible to direct investments towards the most innovative sectors. They can then deduce that “sectoral policies, adapted to the specificities of each sector (development periods, capital intensity, structure of the sector, etc.) are therefore essential” (ibid., p. 43). It then appears once again that the fundamental difficulty facing any policy in favor of industry lies not so much in the dilemma, imposed by the principles of competition policy, between a vertical industrial policy and a horizontal innovation policy, but in the development of a “matrix” policy, making it possible to combine these two orthogonal perspectives. However, this orthogonality, to which the concept of the innovation ecosystem is supposed to respond at least in part, is only an illusion mainly due to the way in which industry is approached by economic theory and the implications that flow from it. To dispel this illusion, it is necessary to specify exactly what the term “industry” covers. 3.2. The nature of the industry Talking about reindustrialization should logically require agreement on the meaning of the word “industry”. Such a reflection seems all the more essential since the implementation in April 2015 of the second phase of the Nouvelle France industrielle program, entitled Industrie du futur (Industry of the Future), which brought together the 34 previous industrial plans into nine industrial solutions, which are: new resources, sustainable cities, ecological mobility, tomorrow’s transport, medicine of the future, data economics, intelligent objects, digital trust and intelligent food. As Gilles Le Blanc suggested, it is particularly useful to move forward in this reflection on the notion of industry to “begin by freeing oneself from classical representations, largely inherited from history. These readings, which still permeate the vision of the general public as well as that of

108

Innovation and Industrial Policies

political leaders, are based on the figures of the factory, assembly lines and symbolic sectors such as the automobile and aeronautics sectors” (Le Blanc 2012, p. 27). Thus, the definition adopted by INSEE is in line with the classic representations denounced by the author since it indicates that “as a first approximation, economic activities that combine factors of production (installations, supplies, labor, knowledge) to produce material goods for the market”. Apart from its approximate nature, this concept refers to the notion of a market, of which it is interesting to note that INSEE does not propose any definition, but simply adds that “a distinction is generally made between manufacturing industry and extractive industries, but the precise outline of industry in each statistical operation is given by the list of items included in the economic classification to which this operation refers”. This vagueness is all the more problematic as “industrial activities have undergone profound changes over the past three decades” (Le Blanc 2012, p. 27). Observation of industrial reality shows that companies establish very varied relationships between themselves, ranging from subcontracting to partnerships, quasi-integration, cooperation agreements, alliances and networking. These different forms of relationships have not only developed quantitatively in recent years, but have also considerably diversified, since they now affect all sectors of activity involved in the production of goods or services1. However, apart from the industrial relations survey already mentioned (Hannoun and Guerrier 1996) or the inter-company relations survey, the first results of which are presented in Haag et al. (2004), the various forms of inter-company relations are difficult to identify because they are ignored by statistics compiled in “branches” or “sectors”. It should be recalled that, for INSEE2, “a branch (or branch of activity) groups together homogeneous production units, i.e. units that manufacture products (or produce services) that belong to the same item of the classification of economic activity in question”. On the contrary, “a sector includes manufacturing, trading or service enterprises that have the same main activity (according to the classification of economic activity considered). The activity of a sector is therefore not entirely homogeneous and includes secondary productions or services that would fall under items other than that of the sector in question”. 1 A first assessment of this transformation of business relations can be found in (Hannoun and Guerrier 1996). 2 Definition provided by INSEE (www.insee.fr).

Reindustrialization Through Innovation

109

Figure 3.1. Distinction between sector and branch (source: Morvan 1991, p. 487). For a color version of this figure, see www.iste.co.uk/ravix/innovation.zip

In this regard, David Flacher and Jacques Pelletan (2007, p. 14) show that “economists face two fundamental questions when using national statistics: the quality of measurement and the relevance of groupings”. The first question refers to the definition and construction of nomenclatures themselves, but also to “the collection of data and their comparability over time and space” (ibid.). The second question arises from the “assumption that the economist’s conception of the industry overlaps with that of the statistician, which is not guaranteed a priori” (ibid.). However, it is this second question that poses the most analytical difficulties. We will therefore start by explaining the problems with the definition of industry (section 3.2.1), so that we can then address the question of its organization (section 3.2.2). 3.2.1. Problems in defining the industry Traditional representations of industry are most often limited to divisions into sectors of activity that include firms producing the same type of

110

Innovation and Industrial Policies

property. As a result, they reduce business-to-business relationships to mere buy-sell relationships and can therefore provide, at best, only a very partial picture of how industry is organized. This is not a new observation since, as early as 1970, Ronald Coase reported on this difficulty at a symposium organized for the fiftieth anniversary of the creation of the National Bureau of Economic Research (NBER). He noted the following: “What is indeed curious about the treatment of problems of industrial organization in economics is that there are none” (Coase 1972, p. 60). Yet, he added: We all know what it is meant by the organization of industry. It describes the way in which the activities undertaken within the economic system are divided up between firms. As we know, some firms embrace many different activities; while for others, the range is narrowly circumscribed. Some firms are large; others small. Some firms are vertically integrated; others are not. This is the organization of industry or – as it is commonly called – the structure of industry” (ibid.). When we look at the various books on industrial organization, it is striking to note that most of them do not concern themselves with giving a definition of industry and even less with proposing a real reflection on a notion that is supposed to be the subject of study in this field of economic analysis. This particularity is a direct result of the difficulties encountered by economic analysis in understanding industry. Thus, when Alexis Jacquemin discusses this question, he begins by noting that: “Theoretically, it is on the basis of consumer demand and the close substitution that exists, in his view, between products, that industry should be delimited” (Jacquemin 1979, pp. 32–33). However, the use of the conditional shows that such an approach raises many difficulties that would justify the fact that, in the end, “the observer is trapped in the available industrial classifications whose criteria often do not correspond to economic concepts” (ibid., p. 33). Such a remark is surprising to say the least. It suggests that the methods developed by statistical agencies would deliberately ignore the “good criteria” of economic theory. The problem is more certainly of a different nature. If statistical agencies are unable to apply the criteria of economic theory, it is probably more simply because such criteria are inapplicable. Under these conditions, there is a better understanding of the reasons why,

Reindustrialization Through Innovation

111

“generally, industrial classification will depend more on supply characteristics (physical or technological structure) than on demand and the degree of substitutability in the eyes of the user” (ibid., p. 33). This reversal of approach is interesting since it tends to show that the coherence of a particular industry should be sought on the production side and not on the consumption side, thus justifying the need for a reflection on the organization of the industry. Such a perspective is very rarely explored by industrial economists, who most often limit themselves to assimilating the organization of industry to that of the market. In particular, this option was adopted by Jean Tirole in a book entitled The Theory of Industrial Organization, which he published in 1988. In his introduction, he asks himself the question: “Why should one be interested in industrial organization?” and adds: “This question sounds almost silly,” because he specifies: “To study industrial organization is to study the functioning of markets” (Tirole 1993, p. 1). Reducing industrial organization to the functioning of the market in this way solves nothing and only moves the problem along, and as he later points out: The notion of a market is by no means simple. Obviously we do not want to restrict ourselves to the homogenous-good case. If we posit that two goods belong to the same market if and only if they are perfect substitutes, then virtually all markets would be served by a single firm – firms produce goods that are at least slightly differentiated (either physically or in terms of location, availability, consumer information, or some other factor) (ibid., p. 12). These remarks lead him to conclude his introduction by indicating that: For the purpose of the present book, this empirical difficulty of defining a market will be ignored. It will be assumed that the market is well-defined, and that it involves either a homogeneous good or a group of differentiated products, which are fairly good substitutes (or complements) for at least one good in the group and have limited interaction with the rest of the economy (ibid., p. 13). This is surprising. Indeed, what can be the relevance of a theory of industrial organization that explains that an industry is in fact a market, but

112

Innovation and Industrial Policies

that it is impossible to define precisely, or not arbitrarily, a market outside the case of a perfectly homogeneous good, while knowing that products are never perfectly homogeneous? Paradoxically, what Tirole’s book demonstrates is that the analysis of industrial organization still tells us nothing about how industry organizes itself. If he thus confirms, 20 years later, the accuracy of Coase’s observation, he nevertheless suggests that the difficulty encountered by economic analysis in understanding industry comes essentially from its basic assumptions. To understand this, it should first be recalled that in standard economic theory, industry and enterprise have no specificity. This absence is the direct consequence of the assumption that all economic agents are fully informed, which implies that they coordinate through a single institution: the market. In this context, agents have no reason to build particular organizations, called companies, to implement their economic activities and, consequently, the very idea of industry, reflecting interdependent relations between these organizations, cannot have any meaning. The question of the nature of industry is therefore closely linked to that of the nature of the firm. However, to bring the firm to the forefront of economic theory, it is necessary to assume that information is no longer available and free. In this case, the use of market coordination, that is a price system, becomes costly and it may be advantageous to use an organization to coordinate certain transactions at a lower cost than the market. This result, demonstrated by Ronald Coase in 1937, in his famous article on the nature of the firm, makes it possible to define the simplest of the institutional divisions of the coordination work. It is a dichotomous and reversible division between the firm and the market. In this perspective, the firm emerges from the market, which raises the question of its borders, that is, the question of sharing the coordination work between the firm and the market. Coase resolves this question by applying a simple substitution rule that allows it to show that the firm is growing until the cost of organizing an additional transaction within the firm becomes equal to the cost of conducting the same transaction in the market. Although Coase’s analysis succeeds in giving a real theoretical status to the company by differentiating it from the market, it nevertheless leads to a complete disruption of the theoretical reference framework since the coordination of economic activities is now ensured by two alternative and

Reindustrialization Through Innovation

113

competing institutions. This situation explains why, as an extension of this theory, two different avenues of analysis have been opened up. The first aims to restore the coherence of the initial theoretical framework by reviving the idea of a single institution for the coordination of agents. This is the path taken by the new microeconomics, which places the contract upstream of the market and the firm as an institutional method of coordination (Cahuc 1993). If the theoretical framework gains coherence through a return to uniqueness, it is at the cost of a new disappearance of the firm’s theoretical identity and, consequently, of inter-company relations. Indeed, this theory leads to a reductive conception of the firm since it understands it through an analysis of bilateral relations completely independent of each other, generated by phenomena of asymmetry of information. The firm therefore essentially presents itself as a “node of contracts”, that is as a simple juxtaposition of parallel and noncomplementary contracts. Such a conception can only lead to the complete eviction of intercompany relations since the latter are purely and simply conceived as isomorphic to the firm’s internal relations, thus preventing any real study of the industry’s organization. This is particularly evident when trying to account for the phenomenon of subcontracting, for example. While this analytical framework explains the mechanisms that govern the contractual relationship between a principal and a subcontractor, it does not qualify the nature of the subcontracting relationship. The latter loses all specificity insofar as “there is a single coordination mechanism that irrigates all economic activity. Subcontracting is recognized here, but this relationship is no different from other contracts: employer-employees, shareholdermanagers, etc.” (Baudry 1992, p. 873). It is with a similar difficulty that the second path of analysis followed by the neo-institutionalist approach is confronted, when it tries to account for the industrial organization. In its initial version developed by Oliver Williamson (1975), this approach is based on a strict dichotomy between the market and the hierarchy based on transaction cost theory. It therefore treats intermediate modalities of coordination as unstable forms that converge either towards the market or towards the hierarchy. However, the subsequent recognition of the importance of business-to-business relations led Williamson (1985) to modify his initial dichotomous model to adapt it to the treatment of these “hybrid forms” which are now considered as stable forms

114

Innovation and Industrial Policies

of coordination of industrial activities. However, this integration of hybrid forms can only be achieved at the cost of a certain blurring of the firm’s borders, since it is only through an implicit assimilation of the organization to a system of long-term contracts that it is possible to interpose between the boundaries of the hierarchy and the market a whole variety of intermediate forms referring to so many different contracts. By converging in this way towards a purely contractualist approach, the transaction cost savings extended to hybrid forms cannot explain the organization of industry (Quéré et al. 1997). To try to overcome this methodological impasse, it is possible to take a third path that is different from the two previous ones. This is the one suggested by Coase himself, in his article on the nature of the firm, when he makes the following observation: A firm will tend to expand until the costs of organizing an extra transaction within the firm become equal to the costs of carrying out the same transaction by means of an exchange on the open market or the costs of organizing in another firm (Coase 1937, p. 395, our emphasis). He thus envisages a case in which a firm will stop its expansion at a point where the additional transaction will have a cost lower than the market, but equal to the cost of organizing in another firm. It therefore explicitly admits the existence of transactions between two producers, each of which could organize at a cost lower than the “real market costs” (ibid.), that is transactions which are not organized either by exchange on the market or by management in a single firm, but which result from a particular relationship between two firms. Coase then asks himself the question “how to solve this paradox?” (ibid.). The solution he uses is to extend the principle of diminishing returns from internal management within the firm to inter-company relations. He considers, first, that these relationships take place between firms that share the different stages of a production process, and second, that the internalization of all these stages by one of the two firms would entail additional organizational costs. He can then deduce that it is worthwhile to “divide production in such a way that the cost of organizing an additional transaction in each firm is the same” (ibid.).

Reindustrialization Through Innovation

115

However, this solution is not satisfactory, because by using a notion of a production process composed of a series of solidarity stages, Ronald Coase is forced to leave his theoretical reference framework. It moves surreptitiously from a problem of coordinating transactions to a problem of organizing production activities, the logic of which completely escapes analysis in terms of substitution (Ravix 1990, 1998). Although not relevant, this solution nevertheless suggests that resolution of the Coase paradox necessarily requires a return to the Marshallian question of industrial organization. 3.2.2. The question of industrial organization In his Principles of Economics, Alfred Marshall was undoubtedly one of the first economists to distinguish organization among the agents of production and to give it an essential role. Often assimilated to a fourth factor of production, organization appears in fact in Marshall’s work as one component of capital next to another one that is just as essential: knowledge. He considers that “capital consists in a great part of knowledge and organization” (Marshall 1920, p. 138). He also specifies that the organization “has many forms, e.g. that of a single business, that of various trades relative to one another, and that of the State providing security for all and help for many” (ibid.). It seems possible to deduce from this last quotation, but also from some of the ideas developed by Marshall in the fourth part of his Principles, that this notion of organization “plays at three different levels” (Ménard 1993, p. 11). The first level would be the general principle that ensures the unity of an economic and social whole. However, as Claude Ménard points out, “in economics, this organizational principle has a precise expression: it is the division of labor, as it articulates specialized functions” (ibid.). The second level would be that of the organization of industry, which “thus characterizes developed societies, where the activity aims at an efficient use of resources through the combination of the technical division of labor and market expansion” (ibid.). Finally, the third level would be the one that Marshall refers to as the business organization, which refers to “the elementary decision-making unit, essentially firms”, thus raising the question of “the comparative effectiveness of the different organizational forms” (ibid., p. 12). Such an interpretation only partially overlaps with Marshall’s definition for at least two reasons. The first is that, in addition to the three levels,

116

Innovation and Industrial Policies

corresponding respectively to the nation, the industry and the enterprise, Marshall also considers “the organization of the various industries with regard to each other”; this level cannot be reduced to that of the nation or that of industry in general. More precisely, Marshall does not distinguish different “levels” of organization that would fit into each other like Russian dolls, but rather different “forms” of organization that are on the same level: that of the implementation and coordination of production activities. The second reason is that the knowledge dimension is completely ignored, even though it plays a central role in defining the organization. Indeed, if for the English economist, “knowledge is our most powerful engine of production”, it cannot be conceived independently of organization insofar as “organization aids knowledge” (Marshall 1920, p. 138). In this way, the idea emerges that knowledge and organization are two consubstantial dimensions of production that play an essential role in the coordination of productive activities. In this perspective, knowledge does not refer to a simple question of information but to the notion of competence (Loasby 1998) which necessarily accompanies the idea that, by nature, production is a timeconsuming act. Indeed, before we can really produce, we must start by investing and building production capacity. The realization of production also requires the implementation of a number of distinct and complementary operations that require variable and often different time periods. More generally, the act of producing is a long-term process since it is renewed. This temporal dimension of production has two important consequences highlighted by Édith Penrose (1959). On the one hand, since production does not take place instantaneously, it must be organized, which makes it possible to understand why “in an industrial economy founded on private enterprise, the business firm is the basic unit for the organization of production” (ibid., p. 9). On the other hand, since the production activity takes time, it generates uncertainty, which leads to considering the company as an entity that devotes “efforts and resources to speculative activity” (ibid., p. 33). The intrinsic uncertainty of the action of producing then presents two fundamental characteristics that condition the speculative nature of companies’ activity. The first is highlighted by George B. Richardson when he introduces a distinction between two notions of information: “market information” and ‘technical information” (Richardson 1990, pp. 29–30). The first concept of information reflects the idea that companies make their

Reindustrialization Through Innovation

117

decisions independently of each other, even though their respective activities are interrelated. The second concept of information concerns the fact that investment decisions are logically made before production processes are put in place, which prevents companies from foreseeing all their implications and thus generates irreversible phenomena. The second characteristic, highlighted by Penrose, concerns the subjective nature of uncertainty. Indeed, “it is the ‘expectations’ and not the ‘objective facts’ that immediately determine the behavior of a company” (ibid., p. 41). The decision-making process then takes the form of a sequential process of trial and error based on a comparison of predictions with the facts characterizing the environment and which necessarily falls within the historical time frame because “in the last analysis, the ‘environment’ rejects or confirms the soundness of the judgments about it, but the relevant environment is not an objective fact discoverable before the event” (ibid.). Decisions are therefore made on the basis of forecasts that do not focus on the various contingencies that may occur, but on how the decision-maker interprets these contingencies. However, for Penrose, “the ‘expectations’ of a firm – the way in which it interprets its ‘environment’ – are as much a function of the internal resources and operations of a firm as the personal qualities of the entrepreneur” (ibid.). The company’s possibilities for action are therefore not directly subject to the uncertainty of its environment but to the state of its knowledge and, more generally, its particular skills. In other words, the firm’s environment is not by nature fundamentally uncertain. It becomes so because the firm has a purely subjective perception of it that ultimately depends on its own past experience but also on its learning capacities. It is therefore understandable why it is not so much the state of technological knowledge that matters as the fact that production “has to be carried out by organizations with appropriate capabilities, or, in other words, with appropriate knowledge, experience and skills” (Richardson 1972, p. 888). It thus appears that, on a strictly analytical level, the existence of the firm as an organization derives solely from the uncertainty created by the temporality of production. From this perspective, the firm’s behavior consists of managing this uncertainty by mobilizing its skills to develop particular specializations that will condition its place within the productive system. Indeed, “at all times a firm has a foothold in certain types of production and in certain types of market, both of which are called ‘areas of specialization’ of the firm”

118

Innovation and Industrial Policies

(Penrose 1995, p. 109). This concept of areas of specialization brings together two different but closely complementary elements. The first element is the basic production element. Penrose considers that “each type of productive activity that uses machines, processes, skills and raw materials that are all complementary and closely associated in the process of production shall be called a ‘production base’ or a ‘technological base’ of the firm, regardless of the number or type of products produced” (ibid., p. 109). It is important to stress that this notion of basic production, completely separate from the notion of a product, covers in fact all the technical and managerial knowledge that the firm must implement to produce. The second element that completes the firm’s field of specialization is that of a market area. Depending on its technological bases, the firm defines market areas that correspond to the fact that it can sell on several markets, even if it has only one production base. In this sense, “each group of customers which the firm hopes to influence by the same sales program is called a ‘market area’, regardless of the number of products sold to that group” (ibid., p. 110). As a result, it is possible to serve several market areas from the same technological base; but also, from several production bases, it is possible to operate on a single market by breaking it down into several areas. This statement that each company builds its own market is not surprising because, “from the point of view of the firm, demand is highly subjective – it is the opinion of the firm’s entrepreneur” (ibid., p. 85). Companies therefore differ in that, on the one hand, according to their own skills, they develop particular production bases, while knowing that “a movement into a new base requires a firm to achieve competence in some significantly different area of technology” (ibid., p. 110), that is, it must acquire or develop new knowledge to create new skills. On the other hand, since “the appropriate criteria for the delimitation of market areas are different for different firms; the significance of the boundaries lies in the fact that a movement into a new market area requires the devotion of resources to the development of a new type of selling program and a competence in meeting a different type of competitive pressure” (ibid., p. 110). Just as companies organize their production bases, they also and logically organize their market areas; the whole constituting their field of specialization. It is thus possible to verify that the concept of field of specialization developed by Édith Penrose corresponds perfectly to the concept of activities adopted by George B. Richardson since the latter are “related to the

Reindustrialization Through Innovation

119

discovery and estimation of future wants, to research, development and design, to the execution and co-ordination of processes of physical transformation, the marketing of goods, and so on” (Richardson 1972, p. 888). On this basis, Richardson introduces an additional distinction between similarity and complementarity of activities. For him, in fact, similar activities are activities that require the same skills to be implemented and complementary activities are those that represent different phases of the same production process. It is important to stress that, from this perspective, coordination problems stem from constraints specific to the implementation of production processes, which require that the activities that they are composed of be organized in a coherent manner by organizations with the required skills. This confirms the idea that production is necessarily carried out within particular organizations (firms) and that there is no production outside these organizations. This distinction makes it clear that firms will tend to group together similar activities for which they have the required skills. Thus, when activities are both similar and complementary, they are coordinated within the firm by management. Conversely, when activities are “closely complementary but dissimilar” (ibid., p. 892), they must be coordinated ex ante by cooperation agreements between companies because the latter, carrying out different stages of the same production process, must agree in advance to harmonize their production plans. Indeed, Richardson notes: this co-ordination cannot be left entirely to the direction within firms because the activities are dissimilar, and cannot be left to market forces, because it does not require balancing the aggregate supply of something with the aggregate demand for it, but does require the matching, both qualitative and quantitative, of individual enterprise plans (ibid., p. 892). On the contrary, all other activities, which are neither similar nor closely complementary, are coordinated ex post in the market. As a result, it is not the market that fundamentally explains the organization of industry, but the different concrete modalities of inter-firm relations, because they condition the “dense network of cooperation and affiliations through which companies are interconnected”. It is this network that Coase describes as the structure of industry and Richardson describes it in the following terms:

120

Innovation and Industrial Policies

Firm A is a joint subsidiary of firms B and C, has technical agreements with D and E, sub-contracts work to F, is in marketing association with G – and so on (ibid., p. 884). Obviously, this structure of industry is by its very nature evolutionary because it results essentially from companies’ activity and mainly from the choices they make to try to stabilize their environment, by trying to dominate it through the control of their technological bases and their market areas. This behavioral logic forces firms to adopt a diversification strategy by building their own field of specialization, because “diversification and expansion based primarily on a high degree of competence and technical knowledge in specialized areas [...] together with the market position, it ensures it is the strongest and most enduring position a firm can develop” (Penrose 1995, p. 119). However, it should be pointed out that the term “diversification” should not be considered in its static dimension referring to a degree of diversification (a firm would be more or less diversified according to the number of products making up its production range), but in the dynamic sense of diversification of productive activities. Indeed, “not only is a comparison of the ‘extent of diversification’ of different firms likely to be meaningless in itself, but statistical studies of the number of different ‘products’ produced by firms are also of very limited usefulness”, because, Penrose points out, “the number of products produced by a firm has no general significance [...]. The definition of a product is at best arbitrary” (ibid., pp. 99–107). From this perspective, competition is not only a process of trial and error in discovering knowledge that is objective but hidden (Hayek 1948, 2002), that is, a reality to be discovered before the agents’ action, which does not differ fundamentally from a world of perfect knowledge (Quéré and Ravix 1995). It also appears as a process of knowledge creation because firms, by adopting diversification behavior, do not limit themselves to reacting to the purely subjective perception of their environment but contribute significantly and consciously to creating this environment. Thus, “firms not only alter the environmental conditions necessary for the success of their actions, but, even more important, they know that they can alter them and that the environment is not independent of their own activities” (Penrose 1995, p. 42); hence the importance of market information, which Richardson believes covers what other firms do. This principle of diversification, by animating the competitive process, is therefore at the root of the cumulative movement which, by widening the division of labor, generates future prospects for specialization.

Reindustrialization Through Innovation

121

Two main phenomena explain why this movement is self-sustaining. The first is clearly identified by Allyn Young when he notices the following: not only new or adventitious elements, coming in from the outside, but elements which are permanent characteristics of the ways in which goods are produced, make continuously for change. Every important advance in the organization of production, regardless of whether it is based upon anything which, in a narrow or technical sense, would be called a new ‘invention’, or involves a fresh application of the fruits of scientific progress to industry, alters the conditions of industrial activity and initiates responses elsewhere in the industrial structure, which in turn have a further unsettling effect (Young 1928, p. 533). The second phenomenon stems from the fact that “even when a firm exploits to the fullest possible extent the opportunities for monopolistic gain available to it, the protection afforded, though often extensive, can neither be complete nor absolutely certain” (Penrose 1995, p. 113). The explanation for this apparent paradox is simply that, as Richardson points out, the traditional analysis ignores the temporal nature of competition. Indeed, after having shown that “increasing returns may not lead to market concentration but to specialization and interdependence”, Richardson asks himself why “will we not find, at the end of the road, precisely the state of monopolistic competition described by Chamberlin?” However, the only possible answer to this question is that, quite simply, “the end of the road can never be reached” (Richardson 1975, p. 357). In other words, the behavior of firms, which characterizes the functioning of competition and conditions the division of labor, constitutes the very essence of economic dynamics. The latter prevents stable or relatively stable positions being perpetuated at any time because it constantly generates new possibilities of the division of labor that companies will try to exploit. Such a conception of competitive dynamics in terms creation processes makes it possible to renew the way we structuring of the industry and the role that companies play reading also offers the possibility of conceiving questions economic role of the State in a different way and thus answers.

of knowledge understand the in it. This new relating to the providing new

122

Innovation and Industrial Policies

3.3. Towards a renewal of State intervention The recurring problems encountered by French industry are generally interpreted today as an indication of the difficulties it faces in gradually moving from an economy of catching up or moving towards the technological frontier to an economy of innovation and knowledge, the purpose of which is to succeed in moving this technological frontier. This idea is largely taken up by recent reports and in particular that of Jean-Lou Blachier (2017), from which the title of this chapter is taken, to argue the need to articulate a genuine renewed industrial policy to innovation policy: Although the PIA, the creation of competitiveness clusters and the establishment of the BPI are almost unanimously welcomed, French industrial policy is nevertheless often perceived by companies as unstable, unclear and insufficiently prioritized. Pursuing a multitude of objectives, it results in a sprinkling of massive public resources on too many targets (Blachier 2017, p. 5). More precisely, on the basis of the companies interviewed as part of its study, this same report considers that France should invest more in certain innovative sectors since “the priorities of Nouvelle France industrielle articulated around the industry of the future and the nine Industrial Solutions were approved by almost all the companies met, which nevertheless wished to associate them with other public industrial policies not covered by the NFI (defense, aeronautics, farm and food, energy, etc.)” (ibid., p. 6). The novelty of this speech is twofold. On the one hand, it lies in the assertion that the suffocation of any industrial policy by competition policy only makes it possible to take advantage of the best opportunities of a given situation, that is on the technological frontier, but not to move it. France’s industrial difficulties would thus come from the fact that it has not been able to find ways to take full advantage of the Third Industrial Revolution (electronics and information technology) and that it must now participate in the next revolution, if it does not want to decline, i.e. that of artificial intelligence, robotics, the Internet of Things, autonomous vehicles, 3D printing, nanotechnologies, bio-technologies, energy storage and quantum computing, to use the fields identified by Klaus Schwab, founder and president of the World Economic Forum, in his 2017 book on the fourth industrial revolution. On the other hand, this discourse has been largely

Reindustrialization Through Innovation

123

reinforced by new developments in economic theory that take the opposite approach to neoliberal economic analysis, denouncing both the “Washington consensus” and the presumption of market efficiency. This new perspective involves rediscovering the virtues of industrial policy (section 3.3.1) and the emerging industry argument (section 3.3.2). 3.3.1. The new virtues of industrial policy The renewal of economic thinking in favor of industrial policy finds its first important developments with Dani Rodrik (2008) who, for more than 10 years, has defended the need for such policies on the basis of two general arguments. The first argument is based on the following observation: “When you scratch beneath the surface of any successful new industry, there is usually some form of State aid” (Rodrik 2010, p. 2). To justify this, he points out, as examples, the central role of the American Department of Defense in the start-up of Silicon Valley or the fact that the American federal government is by far the largest venture capital investor in the United States. Industrial policy can thus be a tool that generates economic benefits, if it facilitates information flows, reduces coordination costs, promotes learning externalities, and creates economies of scale and agglomeration. The importance of industrial policy is even greater if, as it does, it is believed that it is thanks to manufacturing industry that the middle classes exist and thrive in all countries of the world. He points out, in this regard, that: “The presence of a manufacturing industry could ultimately be a fundamental condition for a country’s democratic vigor” (Rodrik 2011, p. 1). Its second argument is based on the idea that there is no a priori distinction between industrial policy and education, health, insurance or macroeconomic stabilization policies (Rodrik 2008, pp. 1–2). All these policies are designed by politicians who can be corrupted or manipulated by lobbies and are implemented to respond to market failures by officials who do not know how to clearly identify and evaluate them. In other words, since the problems raised by any State intervention are not specific to industrial policy, it should not be treated differently, that is, more skeptically than other public policies. Also in the field of industrial policy, understood in the broad sense of “policies that stimulate specific economic activities and promote structural change” (ibid., p. 3), the question is therefore not “why” but rather “how” it should work. In this perspective, three general principles must be respected.

124

Innovation and Industrial Policies

First, “an industrial policy is more a state of mind than a list of specific policies” (Rodrik 2010, p. 2). The main objective must be to build trust between the public and private sectors in order to have the right information to determine where market failures lie, where the barriers to structural change are and where externalities are. Industrial policy is thus first and foremost a process of discovery. Second, “an industrial policy must use both carrot and stick” (ibid., p. 3), which implies that incentives must be temporary and conditional on results achieved. Industrial policy thus consists of clearly determining ex ante the evaluation criteria that will make it possible to decide whether or not an aid is a success. The idea being, as in the field of pharmaceutical research, that success or successes more than compensate for failures (Rodrik 2008, p. 22). Third, “the actors of an industrial policy must remember that its purpose is to serve society as a whole, not just the bureaucrats who manage it or the firms that benefit from it” (Rodrik 2010, p. 3); in other words, industrial policy actors must be accountable to all members of society, that is, they make the process and outcome transparent by explaining how decisions are made and why certain activities or companies are supported (Rodrik 2008, p. 23). The economist can then conclude that “what determines the success of an industrial policy is not the ability to distinguish winners, but the ability to separate themselves from losers – a less insurmountable requirement” (Rodrik 2010, p. 3). The same perspective is taken by James Robinson (2010), whose analysis is similar to that developed by Rodrik on two points: on the one hand, there is a theoretical basis for thinking that industrial policy can play a role in development and, on the other hand, there are real cases where this is indeed what has happened historically. The originality of James Robinson’s analysis comes from his belief that what explains the success or failure of industrial policy lies in the “political economy issues related to the implementation of these policies” (Robinson 2010, p. 21). According to him, what justifies the difference in the results of industrial policies is that they are not homogeneous, but of different types. To be successful, industrial policy “requires an understanding of the political balance of society, its actors and their interests, political institutions, de facto and de jure powers, and how all these elements interact” (ibid., p. 42). Thus, to succeed, an industrial policy requires both a good diagnosis of the productive situation and appropriate tools, but also that those who have the responsibility and political power to promote growth do not block it.

Reindustrialization Through Innovation

125

Taking a more international economic approach, Ann Harrison and Andrés Rodriguez-Clare (2010) argue for a “soft” industrial policy for developing countries, the objective of which would be to create a process of interaction between government, industry and private organizations to collaborate on interventions that directly increase productivity. Thus, instead of using price-distorting tools (taxes, export subsidies, etc.), the authors consider that these countries should implement programs and scholarships to increase worker training, encourage the adoption of new technologies, and improve regulation and infrastructure. This would reduce both corruption and rent-seeking, and it would also be compatible with international trade and investment agreements. Starting from the idea that the main lesson of the great recession of 2008– 2009 was to show that markets were not necessarily efficient, Joseph Stiglitz, Justin Lin and Célestin Monga (2013) succeed in proving the need to reconsider many of the basic ideas of contemporary economic thinking. Thus, they propose to rethink industrial policy in favor of developing countries, which had officially disappeared as a result of the Washington consensus. It should be recalled that the term “Washington Consensus” originated in an article by John Williamson (1990), which presents a set of 10 liberal measures to reform the economies of developing countries and appears to be a consensus among the various American and international institutions based in Washington, such as the International Monetary Fund and the World Bank in particular. According to Stiglitz et al., this review is necessary for two reasons: first, because academic research has highlighted many deep market failures and, second, because all governments, of both industrialized and emerging countries, still have some form of industrial policy. By way of illustration, they mention the American banking sector, where the central bank lends to banks at an interest rate of 1%, allowing them to buy US Treasury bonds earning them 4%, leading the government to subsidizing banks of more than $30 billion each year, which is “much more than any government in a developing country will ever give to an industry” (ibid., p. 8). Extending this analysis, Justin Yifu Lin (2013) proposes to revisit industrial policy in the context of a “new perspective of structural economy”. His analysis is organized around the idea that: The optimal industrial structure of an economy, i.e. the one that will make it the most competitive on domestic and international

126

Innovation and Industrial Policies

markets at a given time, is endogenous to its comparative advantages, which in turn are determined by the specific endowment structure of the economy at that given time (Lin 2013, pp. 59–60). The central question then is how to ensure that a country adopts a growth model that is consistent with its comparative advantages, that is, its “evolving potential of its endowment structure” (ibid., p. 62). The role of the State must then be that of a “facilitating agent” to support the “process of modernization and diversification of its industries through measures to inform, coordinate and compensate for external effects” (ibid., p. 63). Industrial policy then aims to “facilitate the development of its private sector by exploiting the country’s comparative advantages and building on the advantages that new entrants will bring to these markets” (ibid., p. 76). In line with Dani Rodrik’s analyses, these various authors thus consider that the question is no longer whether or not to make an industrial policy, but rather how to make a good industrial policy, that is a policy that accelerates the structural transformation of the economy in order to promote growth. In particular, they develop the idea that externalities of learning and discovery offer a much more robust argument for the thesis of support for infant industries than the conventional argument. 3.3.2. Rediscovering the argument for infant industry The fundamental hypothesis of the thesis defended by Joseph E. Stiglitz and Bruce C. Greenwald (2014) is as follows: If it is true that productivity is the result of learning and that productivity increases (learning) are endogenous, then a focal point of policy ought to be increasing learning within the economy (Stiglitz and Greenwald 2014, p. 23). They conclude that the State therefore has an essential role to play in encouraging learning and fostering the emergence of an innovation economy. Thus, one of the major objectives of economic policy must be to create a learning society or what they call, in the title of their book, “the knowledge society”. Their rejection of the neoliberal doxa is therefore total since they completely dismiss the idea that, in general, markets would be efficient:

Reindustrialization Through Innovation

127

The followers of neither the Arrow-Debreu or the Schumpeterian traditions have succeeded in remedying the obvious lacuna in their analyses: the former have not shown that competitive markets with innovation are Pareto efficient, and the latter have not been able to show that Schumpeterian competition would ensure economic efficiency even in the production of innovations. The reason that they failed is simple: there is, in fact, no presumption that markets where innovation is endogenous are efficient (ibid., pp. 147–148). More generally, Stiglitz and Greenwald consider that innovation does not always improve the well-being of nations. They note that: “Many of the financial innovations directed at circumventing regulations intended to enhance macroeconomic stability contributed to the instability of the economy and played an important role in the 2008 crisis” (ibid., p. 149). In an innovation economy, there is not necessarily a match between private pay and social returns, so it is impossible to assume that markets can be efficient. On the contrary, it must be acknowledged that public authorities must intervene to correct their failures. This approach then leads the authors to question the principle of free trade, which is undoubtedly the most fundamental dogma of liberalism, and more generally of economic theory, since: “Indeed, in some circles, opposition to free trade would be grounds for taking away one’s certification as an economist” (ibid., p. 425). Stiglitz and Greenwald, on the other hand, believe that “it is desirable for governments to intervene in the market, to encourage, possibly through trade protection, sectors in which there is more learning and more learning spillovers” (ibid., p. 210). They thus rediscover the spirit of educational protectionism, initially inspired by Friedrich List in his book entitled The National System of Political Economy (1856). List’s fundamental idea was that the only way for a country to develop a new industry, subject to strong international competition, was to protect it with customs barriers until it could reach full maturity and thus have the opportunity to compete with rival countries. This thesis is not fundamentally hostile to free trade since the protectionism it advocates is purely transitory. It simply aims to avoid irreversible phenomena of international specialization, generated solely by comparative advantages. However, Stiglitz and Greenwald add an additional dimension to this infant industry argument by focusing on the phenomena of innovation,

128

Innovation and Industrial Policies

knowledge and, above all, learning. They prefer to speak of the “infant economy argument” in order to not focus their analysis on particular sectors but on the economic system as a whole; this leads them to propose a complete renewal of the purpose of public intervention. They consider that: in this perspective, industrial and trade policy is not focused on picking winners, though, to be sure, governments do not want to pick losers. Nor is it predicated on the belief that government can do a better job than the private sector in picking winners. It is based on the notion that learning involves spillovers (externalities) that will be imperfectly internalized in a market economy. Industrial and trade policies are concerned with identifying sectors or industries (firms, areas of innovation) which would generate large externalities or where the returns that could be appropriated by the innovating (learning) firm are a fraction of societal benefits. Governments in many countries have, in fact, done a credible job in making these selections, and our societies have benefited greatly as a result (ibid., p. 214). The change in perspective introduced by Stiglitz and Greenwald is somewhat revolutionary since they no longer justify industrial and trade policies by negative market failures (negative externalities), but by positive market failures (positive externalities and public goods), passing through the need for the State to permanently maintain the learning process, because it is this process that truly conditions the functioning of economic activity. It is no longer a question of resource allocation that the market makes that takes precedence, but innovation or the creation of knowledge supported by the State. This radical change of perspective is also reflected in the various works carried out by Philippe Aghion, in collaboration with other authors. Starting from the idea that France is now facing the end of a catching-up phase, corresponding to European integration, and the transition to an innovation economy, he considers it necessary to rethink industrial policy. The new perspective he develops is based on three observations (Aghion et al. 2011): first, the more objectively subsidies are targeted on sectors and not on one or more firms, the more growth there is; second, the more competitive or procompetitive the targeted sectors are, the more competition will encourage innovation and thus growth, and finally, public aid has an even greater

Reindustrialization Through Innovation

129

impact on growth if it concerns sectors with a high concentration of skilled labor. In terms of industrial policy governance, Aghion and his co-authors also suggest that the more decentralized a system of public support is, the more effective it is in terms of competition, export competitiveness and innovation. In particular, they consider that sectoral intervention is mainly needed in two areas: renewable energy and the environment, where “the optimal policy for combating climate change is to combine a carbon tax and a policy of subsidizing clean innovations” (Aghion et al. 2014, p. 200); and in the area of financing to counter the existence of credit constraints that “may hinder the reallocation of capital to sectors with high growth potential” (ibid., p. 202). Thus, as these authors acknowledge: there are activities – usually high-tech sectors – that generate spillovers to the rest of the economy and whose assets are highly intangible, making it more difficult for companies to borrow in private capital markets to finance their growth. It could then be justified to subsidize entry and innovation in the relevant sectors and to do so in a way that ensures fair competition in the sector (Aghion et al. 2014, p. 521). More generally, Aghion et al. (2015) support the idea that industrial policy should not be systematically opposed to competition policy. In this perspective, the “rehabilitation” of sectoral policies, targeted at certain sectors and not at particular firms, is based on the theoretical argument that “targeted subsidies can be used to encourage several firms to operate in the same sector and that the more competitive the sector, the more it will encourage firms to innovate” (Aghion 2014, p. 522). If innovation was already the driving force behind growth, it is now also becoming the driving force behind competition. 3.4. Conclusion All these recent theoretical developments show that industrial policy, competition policy and innovation policy can no longer be considered as alternatives, but rather as supplementary. Nevertheless, it is necessary to go back in part to the theoretical justifications for the rediscovery of the virtues of industrial policy because

130

Innovation and Industrial Policies

two ambiguities arise. The first concerns the choice of paradigm made by authors such as Joseph Stiglitz and Bruce Greenwald, who nevertheless place their reflections on innovation processes in “broader, evolutionary terms” (Stiglitz and Greenwald 2014, p. 149). They write in their book: Though the results of our analysis differ distinctly from that of the standard neoclassical analysis, the tools of analysis we use in most of this book are standard (ibid.). This ambiguity suggests that the results obtained would not be fundamentally linked to the methodological choice made, but to the models used, and consequently that the standard approach and the evolutionary approach would be complementary and not alternative. However, it should be recalled that it was from the 1980s onwards that much work on technological innovation contributed to the emergence of an evolutionary current that clearly differed from standard economic theory (Nelson and Winter 1982; Dosi et al. 1988; Saviotti and Metcalfe 1991). Borrowing the concept of creative destruction from Schumpeter (1942), this approach rejects the idea of a general equilibrium, and instead emphasizes the dynamics of qualitative change, the importance of the role of uncertainty and the heterogeneity of economic agents. The second ambiguity is linked to the first and “new” nature of the results, which affirm the compatibility between vertical industrial policy and horizontal innovation policy. It is indeed possible to note that, as Pier Paolo Saviotti already noted in 1995: the nature of competition between firms has undergone very strong changes since the 1980s. The most important dimensions of these changes are globalization, liberalization, the emergence of several new technologies and inter-institutional collaborations. These changes in the context of competition can have very profound implications for industrial policies (Saviotti 1995, p. 199). On the basis of this observation, he could diagnose a renewal of industrial policies, based on the point of view of evolutionary analyses, while underlining the profound changes that must take place in the very nature of these policies. The novelty emerged from the following fact:

Reindustrialization Through Innovation

131

Traditional industrial policies have always been based on the assumption that the state of equilibrium, corresponding to perfect competition, is the most effective in the economic system, and that the role of industrial policies is to return the system to that state when it moves away from it. While in open systems, structural changes occur as the system moves away from balance, the role of industrial policies is diametrically opposed. Here, it is a question of promoting the distance of the system from equilibrium (ibid., pp. 204–205). There would therefore be a profound rupture between the neoclassical and evolutionary approaches, which can only be found in the justification of public policies. Paradoxically, if economic and political discourses seem to be more in tune today, this is not fundamentally the result of the awareness of an important transformation of industrial realities, nor of the emergence of new paradigms in economic science, but it is the consequence of an evolution of standard economic theory which, now that it has succeeded in taking into account certain aspects of the phenomena of innovation, imperfect competition and endogenous dynamics, opens the possibility to analyze the determinants of industrial policies in a different way. Once again, it is the formal coherence of economic theory that justifies policies to support industry, not the empirical relevance of industrial issues that determines economic theory.

Conclusion

One of the current main challenges of industrial and innovation policies is evaluating their impact. This examination is in line with the recurrent concern of reducing public spending in order to achieve budgetary savings. This is a very old concern, inherent in all forms of public action, but its modern version can be traced back to the experience of the Rationalization of Budgetary Choices (Rationalisation des choix budgétaires, RCB) which was introduced in 1968 and abandoned in the 1980s. This rationalization was to be achieved by monitoring the results of administrative action through cost-effectiveness studies. As Antoine Bozio (2014) points out, an equivalent objective was assigned in 2001 to the Organic Law on Finance Laws (Loi organique relative aux lois de finances, LOLF), then to what has been called, since 2007, the General Review of Public Policies (Révision générale des politiques publiques, RGPP) and, more recently, in 2012, to the Modernization of Public Action (Modernisation de l’action publique, MAP). This perpetual currency could lead to considering the evaluation as a recurring phraseology of government action: invoked, in turn, as a tool for legitimizing public action, a means of democratizing political life, a response to socio-technical controversies, a lever for sectoral action by ministries and a vehicle for public performance, the evaluation was mobilized on a wide range of issues. The latter constitutes scenes where different groups of actors with heterogeneous skills and resources have been deployed, but what they have in common is that they have not succeeded in imposing their vision of

Innovation and Industrial Policies, First Edition. Joël-Thomas Ravix and Marc Deschamps. © ISTE Ltd 2019. Published by ISTE Ltd and John Wiley & Sons, Inc.

134

Innovation and Industrial Policies

evaluation and thus provide the basis for a “French approach” to evaluation (Lacouette-Fougère and Lascoumes 2013, p. 860). However, in this area, a distinction must be made between two different conceptions of evaluation. Thus, in the preamble to a report on the evaluation of public policies and actions (Viveret 1989), Lionel Stoleru pointed out that if France has a long tradition of monitoring the use of public funds [...] it must be noted that budgetary surveillance has only distant links with economic evaluation. The first covers the lifetime of a project, whereas the second is an annual one. The first extends to all income and expenditure, whereas the second is only concerned with expenditure and the budgetary fraction of this expenditure (Stoleru 1989, p. 5). It is this idea of a necessary economic evaluation that will nevertheless become established. Didier Migaud, first President of the Cour des comptes (Court of Auditors), recalls that Michel Rocard, then Prime Minister, referred, in a circular of February 23, 1989 on the renewal of the public service, to a “duty of evaluation”, the justification for which is mainly “a concern for transparency in the use of public funds, which is one of the components of democracy” (Migaud 2013, p. 850) and which, as a result, requires that any public action be evaluated. He also stated that “this requirement is contained in article 15 of the Declaration of the Rights of Man and of the Citizen, which states that ‘the Society has the right to hold any public official accountable for its administration’” (ibid.). The democratic nature of the evaluation of public actions is widely developed and supported by the Economic, Social and Environmental Council (Conseil économique, social et environnemental, CESE) in its report Promouvoir une culture de l’évaluation des politiques publiques, presented by Nasser Mansouri-Guilani (2015). In particular, this emphasizes the idea that, while evaluation aims to facilitate and improve political decision-making, it is mainly, for the actions of the State, about accountability to citizens, and is therefore essential to ensure confidence in political action. However, such a concern for transparency is in general, difficult to implement, because of the complexity of public policies, which has

Conclusion

135

increased over time, but also because of the constant increase in the number of actors involved. However, most of these economic and social interventions generate “windfall effects, the measurement of which is essential to assess the effectiveness and efficiency of a policy” (Migaud 2013, p. 850). It is therefore not surprising that “the Court (of Auditors) often finds that public policies are poorly targeted, due to a lack of prior analysis and sufficient reflection” and Didier Migaud can then deduce that “a public policy approach is necessary to identify all the actors involved in it and measure its results on society and the economy” (ibid.). If we compare the French situation with that of the Anglo-Saxon countries or with that of other European countries, we must recognize that “France did not join in the evaluation of public policies until late” (Fouquet 2011, p. 10). This culture only really began to be expressed in the 1990s, following the recommendations of the report on the evaluation of public policies and actions (L’évaluation des politiques et actions publiques (Viveret 1989)). Indeed, it was the decree of January 22, 1990 on the evaluation of public policies which, with the establishment of an interministerial evaluation mechanism – based on a Conseil scientifique de l’évaluation (CSE), replaced in 1998 by the Conseil national de l’évaluation (CNE) – made it possible to create a genuine institutional mechanism dedicated to the evaluation of public policies. In this new system created by the decree of November 18, 1998, the CNE presented itself as a kind of administrative authority and the Commissariat general du Plan (General Planning Commission) retained an operational role and the prerogatives entrusted to it by the 1990 decree. In particular, it can be seen that “the decree enriches the theoretical role of the General Planning Commission, which is recognized as the administrative institution that acts as the relay for the inter-ministerial evaluation system for public policies then in place” (Bourdin et al. 2004, p. 184). Following the replacement in 2006 of the Commissariat général du Plan by the Centre d’analyse stratégique (CAS), then, in 2013, by France Stratégie, the latter is now responsible for evaluating public policies. In the fields of innovation and industry, it is more precisely the Commission nationale d’évaluation des politiques d’innovation (CNEPI), set up under France Stratégie in June 2014, which is responsible for carrying out evaluations of these policies.

136

Innovation and Industrial Policies

When it comes to assessing public action, the very idea of evaluation may have been perceived negatively, because it would transform a political debate into a purely technical discussion (Fouquet 2013) or, more critically, “as the domination of bureaucratic expertise aimed at replacing democratic choices” (Bozio 2014, p. 62). This latter interpretation may be justified, but it may also be underlined by the fact that an evaluation procedure essentially uses quantitative criteria since it must be able to measure the concrete effects of a policy. However, such an approach would inevitably lead to transforming any qualitative determination into a quantitative one, through the generalization of costing and a kind of numerical scholasticism. The evaluation, which is always subjective and relative, seeks to hide behind a shoddy mathematics. This is the reason for the generalization of the costing: it is convened to give a luster of objectivity to what is often an act of power (Zarka 2009, p. 117). An equivalent point of view is developed by Angélique Del Rey, in her book La tyrannie de l’évaluation (2013), which also denounces the fact that “in the name of performance, we create a measure that takes off from reality and only ‘measure’ the ability of the real to conform to the measures” (Del Rey 2013, p. 46). While there is no doubt that the costing methods, on which most evaluation procedures are based, raise legitimate questions, it must be acknowledged that these methods have improved considerably and diversified in recent years. Without going into detail on the different evaluation procedures and techniques, which are summarized in a note from the Conseil d’analyse économique (2013), two main points should be recalled. On the one hand, “a distinction is generally made between ex ante and ex post evaluation. Ex ante evaluation is carried out before the introduction of a policy and consists of analyzing its potential effects; ex post evaluation aims to measure its real impact after it has come into force. In both cases, the effectiveness of the policy studied is measured by comparing its costs and benefits” (Bozio 2014, p. 65). On the other hand, since the 1990s, the evaluation has increasingly used econometric methods. Although there is not yet a consensus in this area, it can be considered that there are essentially two different approaches. The first, which is described as “structuralist”, is based on the explanation of a complete model of economic agents’ behavior. As Pauline Givord (2014, p. 3) shows, “this type

Conclusion

137

of approach is particularly useful ex ante, to predict the outcome of a policy that has never been applied before”. The second approach, which is more “empirical”, consists of “directly testing the effect of a measure on what interests the econometrician, without necessarily explaining all the mechanisms at work” (ibid.). The Commission nationale d’évaluation des politiques d’innovation (CNEPI), subscribes to this second approach; in January 2016, it published its first report on fifteen years of innovation policies in France (Harfi and Lallement 2016). The justification for this study was the observation of the great complexity of the innovation support system, which had gradually been put in place over time, and therefore of the difficulty in understanding how it works. The report shows that, between 2000 and 2014, public financial support for innovation doubled in constant euros and that this commitment to innovation continued despite political changes. The report also underlines that “this effort has been put at the service of a national ambition: to place the French economy at the technological frontier in a sustainable way, to ensure its competitiveness by increasing the range of goods and services produced” (ibid., p. 7) and echoes Philippe Aghion’s formula, according to which we must complete the transition from our old imitative economy to an innovation economy. Four main observations emerge from this evaluation report. The first is that total innovation aid amounted to “€10 billion in 2014, or half a percentage point of GDP” (ibid., p. 8). This is a considerable amount, which is provided mainly by the State and its operators (87.2%), then, in smaller proportions, by the regions and other local authorities (7.4%) and finally by Europe (4.5%). The second observation is that the modalities of innovation support are characterized by “a multiplicity of objectives, a profusion of instruments and an instability of mechanisms” (ibid.). More specifically, the report notes the existence of inevitable redundancy since “in 2000, the State and its operators managed nearly 30 national schemes. Their number has increased to 62, to which must be added those managed by local and regional authorities” (ibid.). The third observation concerns the evolution of the support modalities, which show a significant shift from direct to indirect aid and more particularly towards “tax incentives, i.e. essentially the research tax credit (CIR), [which] now represents, with €6.4 billion, more than 60% of total support, against 17% in 2000” (ibid.). The fourth observation is that, during the fifteen years covered by the report, “a major institutional reorganization has been carried out with the establishment of

138

Innovation and Industrial Policies

two major players: the Commissariat général à l’investissement (CGI), which manages the Programmes d’investissement d’avenir (PIA), and the Banque publique d’investissement (Bpifrance), which supports and finances companies’ innovation efforts” (ibid., p. 9). On the basis of these findings, the report identifies five main objectives that correspond to the general objectives of innovation policies and that seem unquestionable. These are, respectively, to increase private R&D capacities, increase the economic impact of public research, develop cooperation projects between actors, promote innovative entrepreneurship and finally support the development of innovative firms. In its conclusion, the CNEPI report questions, among other things, two characteristics of innovation policy in France. On the one hand, he notes that it is most often the absence or insufficiency of private investment that justifies public intervention. Therefore, “it will be necessary to identify the main reasons for these deficiencies and to examine whether they cannot be addressed directly at source, rather than through a multiplication of palliative mechanisms” (ibid., p. 88). On the other hand, it raises the question of the role of nontechnological innovation (commercial, organizational, social or designrelated) in relation to technological innovation: While the positive externalities generated by technological innovation are well documented, other types of innovation play a key role in the new business models at the heart of the growth of advanced economies (ibid., p. 89). In parallel with this first evaluation, France Stratégie also approached Philippe Maystadt, former Director of the European Investment Bank, to chair a committee responsible for conducting a mid-term review of the Future Investment Program (Maystadt 2016). The committee’s report notes that “overall, the PIA is an original initiative that has produced positive effects, both quantitative and qualitative” (ibid., p. 25). In particular, it notes a number of strong points: the relevance of the proposed strategic priorities, the principle of excellence in the choice of actions and projects, the use of independent juries, but the main point is that “financial monitoring by the IGC is robust; the allocation of resources per action is transparent, with regular updates of the amounts committed, contracted and disbursed, presented in particular to the relevant committees of Parliament” (ibid., p. 26). This review committee also notes at mid-term that “during its implementation, the PIA has been subject to abuses that have partially

Conclusion

139

distanced it from its initial objectives” (ibid.), mainly by adopting a very broad concept of the notion of future investment, both in terms of defining actions and choosing sectors and selection criteria. Two of them deserve to be clarified. On the one hand, the committee observes “an extension of actions to fields far from investment in knowledge and innovation and not among the priorities defined by the Juppé-Rocard commission” (ibid., p. 33). On the other hand, it indicates that many sectoral actions have been developed, probably to allow a greater visibility of certain priorities such as health, sustainable development or digital development, but also for reasons of redeployment and budgetary substitution. However, the committee considers that “certain actions or funding modalities are incompatible with the initial doctrine of the PIA (for example, subsidies to certain sectors or the (re)financing of existing structures)” (ibid.). After having established these comprehensive diagnoses of the overall evolution of innovation measures, France Stratégie issued, the following year, through the Commission nationale d’évaluation des politiques d’innovation (CNEPI), an opinion on competitiveness cluster policy (Harfi and Lallement 2017), which reviews ten years of this policy, based more specifically on a study carried out for France Stratégie by Haithem Ben Hassine and Claude Mathieu (2017), entitled Évaluation de la politique des pôles de compétitivité: la fin d’une malédiction? This CNEPI opinion echoes the finding of the France Stratégie study that “the impact on business R&D is positive, with a substantial leverage effect” (Harfi and Lallement 2017, p. 2). Indeed, the companies that have joined the clusters, unlike equivalent companies that have not, have increased the selffinancing of their R&D activities, beyond the public aid (direct and indirect) received. Thus, “for an additional euro of public funding, it is on average nearly three euros – including about two euros from its own funds – that a company that is a member of a cluster incurred in R&D expenditure in 2012” (ibid.). The CNEPI also notes that “this leverage effect of public financing is an important result that has never before been identified by the studies on French cases, and which has only rarely been highlighted abroad” (ibid.). This positive observation is offset by the fact that, “given the data available, which do not yet extend beyond 2012, i.e. three years after the end of the first R&D projects financed, no significant effect is detected on the performance downstream of R&D (number of patents filed, turnover, exports, employment, labor productivity, etc.)” (ibid.).

140

Innovation and Industrial Policies

The situation therefore appears to be mixed since, until 2012, the knockon effect on R&D investment does not seem to have translated into better performance for the firms concerned. However, this situation could be explained by the lack of data, due to the time required for the maturation and development of some industrial projects. In addition, account must be taken not only of the halving, during the period analyzed, of the budgetary resources allocated to this competitiveness cluster policy by the State, but also of the establishment and development of other mechanisms with equivalent objectives such as the Future Investment Program (Programme d’investissement d’avenir, PIA) or the Research Tax Credit (Crédit d’impôt recherche, CIR). The CNEPI concludes that competitiveness clusters can be a very useful instrument to stimulate innovation and cooperation between the various actors, but only if the State manages to better define the objective it pursues through this policy. Depending on the type of option chosen, two approaches are possible. The first consists of concentrating State support on the most structuring poles, in line with the “nine industrial solutions” that form the basis of the industrial policy guidelines defined within the framework of L’industrie du futur, the second phase of La Nouvelle France industrielle, launched in 2015. The second logic, more consistent with the promotion of open innovation and the renewal of the productive fabric, consists, on the contrary, of “raising the level of requirements when selecting projects, better defining the criteria for the success of the clusters, and taking greater account of the results of evaluations when periodically reviewing their accreditation” (ibid., p. 4). The aim, therefore, is to encourage the best strategies and give the regions the opportunity to support a greater number of more local centers of interest by using the labeling and financing of projects. But, according to the CNEPI’s opinion, in either case, “the purposes and criteria for using the two State levers of action, namely the labeling of clusters and their public funding, must be more clearly separated” (ibid.). All in all, therefore, while these recent evaluations express generally positive opinions on the various components of industrial and innovation policy which have been in place in France for some 15 years, they nevertheless present two characteristic features that deserve to be highlighted. The first is that these procedures are increasingly oriented towards the use of econometric valuation methods, although these techniques are still difficult to apply. The study by Ben Hassine and Mathieu (2017), by positioning itself in relation to some of these approaches, evokes

Conclusion

141

the main difficulties they raise. However, it remains silent on some of the methodological problems caused by the use of econometric measurements. Among these, the main one is to focus only on formal or technical problems, leaving aside the wide variety of issues raised by the evaluation process itself. However, the latter are analyzed in detail in a report by the Assemblée nationale on public policy evaluation mechanisms, which addresses in particular the problems relating to the various dimensions and characteristic stages of any evaluation, procedure, through the contribution of Maurice Baslé, Jean-Michel Josselin and Benoît Le Maux (2018). Although much progress has been made in this area, an important problem remains, since the question of the relevance of the theoretical reference framework is never examined. Indeed, if, as Annie Fouquet (2011, p. 10) notes, “any policy is based on an implicit theory of social change”, the choice of this theory is not questioned at any time: the evaluation of a public policy is based on a number of criteria, starting with its relevance, which verifies whether the action taken was in line with the problem at hand. Then, it is necessary to measure the coherence of the actions between them and with the means deployed. There are also questions of effectiveness (are the achievements, results and impacts achieved in line with the objectives?) and efficiency (this relates the results achieved to the expenses incurred) (ibid., p. 11). It should be noted that the “relevance” criterion adopted here does not concern the reference theory, but only the adequacy between the policy chosen and the problem at hand, forgetting that the choice of such a policy is nevertheless largely determined by the reference theory. This question of the relevance of the theory appears clearly in relation to competition, since the economic literature offers two distinct conceptions of the idea of competition. In the first, specific to traditional theory, “competition is defined and studied as a state of the industrial system at some point in time” (Gaffard and Krafft 1996, p. 16). On the contrary, in the second, described as “alternative”, competition is considered as a process, which has two effects: The first is that competition is this time an intrinsically temporal phenomenon, which evolves over time. The second is

142

Innovation and Industrial Policies

that a one-off measure of the degree of competition no longer makes any real sense (ibid.). The ambiguities associated with this divide are particularly evident when it comes to talking about competition policy. It is indeed obvious that “competition seen as a process does not have the same implications as competition seen as a state of equilibrium. Short-term and long-term approaches differ, particularly in terms of the relevance of innovation policies” (Encaoua and Guesnerie 2006, p. 294). However, reducing the distinction between state competition and process competition to a simple opposition between the short- and long-term has the effect of completely obscuring the theoretical divide and its implications. The process is reinforced by the attribution of the notion of process competition to “the Neo-Austrian School, which is now a very small minority” (ibid., p. 50), as if such an observation could have any scientific value. However, it makes it possible to avoid the fact that this conception does not only belong to the Hayekian tradition, but also conditions the classical tradition (McNulty 1968) and serves as a basis for the post-Marshallian approach (Penrose 1995; Richardson 1990). We find here, although in a slightly different form, the same observation as the one made previously about the notion of industry (see section 3.2). It is then possible to broaden the scope by noting that phenomena such as competition, industry or innovation, which are, respectively, the subject of, and material for, specific but largely correlated public policies, paradoxically have all the characteristics of a mirage, insofar as they refer to a perception or interpretation of reality. It should also be noted that observations and empirical measures, often put forward to justify their relevance, only reinforce the errors, since we know that statistics can only relate to phenomena that theory can conceive and not to those that it ignores. We are thus confronted with what Jean-Paul Fitoussi calls “the theorem of the lamp post” which refers to the well-known story “of the person who looks for his keys under a lamp post not because he has lost them there, but because it is the only lit place on the street” (Fitoussi 2013, p. 13). Indeed, as he points out, “we are the ones who choose what to shed light on, what phenomena to analyze, what metrics to use, what objectives to pursue” (ibid., p. 12). The crucial question is therefore to know what we are illuminating. In other words, if the definitions used and the choices made are not relevant, the results may be unsuccessful or even incorrect:

Conclusion

In the field of political action, this can have serious consequences, because errors can accumulate – errors in the definition of the objective, to its extent, in the choice of instruments used according to the purposes pursued, i.e. the theory or doctrine that will govern the action (ibid., p. 12). The stakes are therefore high.

143

References

Acemoglu, D. (2009). Introduction to Modern Economic Growth. Princeton University Press, Princeton. Aghion, P. (2014). Commentary: The Case for Industrial Policy. In Creating a Learning Society: A New Approach to Growth, Development, and Social Progress, Stiglitz, J.E., Greenwald, B.C. (eds). Columbia University Press, New York, 517–523. Aghion, P., Howitt, P. (1992). A model of growth through creative destruction. Econometrica, 60(2), 323–351. Aghion, P., Howitt, P. (1997). Endogenous Growth Theory. MIT Press, Cambridge, Massachusetts. Aghion, P., Howitt, P. (2008). The Economics of Growth. MIT Press, Cambridge, Massachusetts. Aghion, P., Boulanger, J., Cohen, E. (2011). Rethinking industrial policy. Bruegel Policy Brief, 4, 1–8. Aghion, P., Cette, G., Cohen, E. (2014). Changer de modèle. Odile Jacob, Paris. Aghion, P., Cai, J., Dewatripont, M., Du, L., Harrison, A., Legros, P. (2015). Industrial policy and competition. American Economic Journal: Macroeconomics, 7(4), 1–32. Arrow, K. (1962). Economic welfare and the allocation of resources for invention. In The Rate and Direction of Inventive Activity, National Bureau of Economic Activity (ed.). Princeton University Press, Princeton, 609–626. Artus, P., Virard, M.-P. (2008). La France sans ses usines. Fayard, Paris.

Innovation and Industrial Policies, First Edition. Joël-Thomas Ravix and Marc Deschamps. © ISTE Ltd 2019. Published by ISTE Ltd and John Wiley & Sons, Inc.

146

Innovation and Industrial Policies

Bacon, F. (1625). Des innovations. Essais de morale et de politique. In Œuvres philosophiques, morales et politiques de François Bacon, Buchon, J.A.C. (ed.). Auguste Desrez, Paris, 491–493. Bagwati, J. (2007). In Defense of Globalization. Oxford University Press, Oxford. Baldwin, R., Martin, P. (2004). Agglomeration and regional growth. In Handbook of Regional and Urban Economics, vol. 4, Henderson, J.V., Thisse, J.-F. (eds). Elsevier North-Holland, Amsterdam, 671–2711. Barro R.J. (1990). Government spending in a simple model of endogeneous growth. Journal of Political Economy, 98(5), 103–125. Barro R.J., Sala-i-Martin X. (1999). Economic Growth. MIT Press, Cambridge, Massachusetts. Baslé, M., Josselin, J.-M., Le Maux, B. (2018). Dispositifs d’évaluation des politiques publiques et des programmes : connaissances de base, choix des méthodes, sociogramme des acteurs et études de cas. Report 771, Assemblée nationale, Paris, 143–206. Baudot, P.-Y., (2011). L’incertitude des instruments. L’informatique administrative et le changement dans l’action publique (1966–1975). Revue française de sciences politiques, 61(1), 79–103. Baudry, B. (1992). Contrat, autorité et confiance : la relation de sous-traitance estelle assimilable à la relation d’emploi ?. Revue économique, 43(5), 871–894. Beffa, J.-L. (2005). Pour une nouvelle politique industrielle. La Documentation française, Paris. Bellon, B., De Bandt, J. (1991). La politique industrielle. In Traité d’économie industrielle, 2nd edition, Arena, R., Benzoni, L., De Bandt, J., Romani, P.-M. (eds). Economica, Paris, 824–846. Ben Hassine, H., Mathieu, C. (2017). Évaluation de la politique des pôles de compétitivité : la fin d’une malédiction ?. Working document, France Stratégie, Paris. Benassy-Quéré, A. (2012). Au secours, notre industrie fiche le camp !. CEPII le blog [Online]. Available: http://www.cepii.fr/BLOG/bi/post.asp?IDcommunique=78. Benassy-Quéré, A., Coeuré, B., Jacquet, P., Pisani-Ferry, J. (2017). Politiques économiques, 4th edition. De Boeck, Brussels. Bentham, J. (1816). Tactique des assemblées législatives suivie d’un Traité des sophismes politiques, vol. 2, J.J. Paschoud (ed.), Geneva. Berger, S. (2013). Making in America. From Innovation to Market. MIT Press, Cambridge, Massachusetts.

References

147

Berthelin, P.-C. (1762). Abrégé du dictionnaire universel françois et latin, vulgairement appellé (sic) Dictionnaire de Trévoux. Libraires associés, Paris. Besnard, L., Bernardini, J.-J., Dejean, E. (2007). Le financement de la recherche d’intérêt général : le rôle des fondations. In La gouvernance de l’innovation : marché et organisations, Boutillier, S., Uzunidis, D. (eds). L’Harmattan, Paris. Betbèze, J.-P., Saint-Étienne, C. (2006). Une stratégie PME pour la France. La Documentation française, Paris. Beylat, J.-L., Tambourin, P. (2013). L’innovation : un enjeu majeur pour la France. La Documentation française, Paris. Bidet-Mayer, T., Touba, L. (2013). À quoi servent les filières ?. Presses des Mines, Paris. Billoret, J.-L. (1989). Condorcet, créateur de la théorie de l’innovation. In Condorcet mathématicien, économiste, philosophe, homme politique, Crépel, P., Gilain, C. (eds). Minerve, Paris. Birch, D. (2001). The Gazelle Theory. Inc. Magazine, 23(7), 28–29. Blachier, J.-L. (2017). Réindustrialiser par l’innovation. Report, Ministère de l’Économie et des Finances, Paris. Blanc, C. (2004). Pour un écosystème de croissance. Report, Assemblée nationale, Paris. Bougette, P., Deschamps, M., Marty, F. (2015). When economics met antitrust: The Second Chicago School and the economization of antitrust law. Enterprise & Society, 16(22), 313–353. Bourdin, J., André, P., Plancade, J.-P. (2004). Évaluation des politiques publiques en France. Report 392, Sénat, Paris. Bourguignon, F. (2012). La mondialisation de l’inégalité. La République des idées, Paris. Bouvier, Y. (2008). Construire l’Europe industrielle par les entreprises. La politique de la concurrence et les fusions industrielles dans les télécommunications européennes des années 1980. Histoire, économie & société, 1, 79–90. Bozio, A. (2014). L’évaluation des politiques publiques : enjeux, méthodes et institutions. Revue française d’économie, XXIX(4), 59–85. Brander, J. (1981). Intra-industry trade in identical commodities. Journal of International Economics, 1(1), 1–14. Brander, J., Krugman, P.R. (1983). A “reciprocal dumping” model of international trade. Journal of International Economics, 15(3/4), 313–321.

148

Innovation and Industrial Policies

Brander, J., Spencer B. (1985). Export subsidies and international market share rivalry. Journal of International Economics, 18(1/2), 83–100. Bréchon-Moulènes, C., (1992). La concession de service public face au droit communautaire. Sirey, Paris. Buigues, P.-A., (2016). Quelle politique industrielle pour éviter le décrochage industriel ?. Working document 35, GREDEG, Valbonne. Cahuc, P. (1993). La nouvelle microéconomie. La Découverte, Paris. Cantillon, R. (1952 [1755]). Essai sur la nature du commerce en général. INED, Paris. Chamberlin, E.-H. (1962 [1933]). The Theory of Monopolistic Competition, 8th edition. Harvard University Press, Cambridge. Chaptal, J.-A. (1819). De l’industrie française. A.A. Renouard, Paris. Clapham, J.-H. (1922). On empty economic boxes. The Economic Journal, 32(127), 305–314. Coase R. (1937). The nature of the firm. Economica, 4(16), 389–405. Coase R. (1972). Industrial organization: A proposal for research. In Policy Issues and Research Opportunities in Industrial Organization, Fuchs, V. (ed.). National Bureau of Economic Activity, pp. 59–73. Coase, R. (1991 [1972]). L’organisation industrielle : un programme de recherche. Revue d’économie industrielle, 58(4), 15–27. Cohen, E. (2012). Stratégie de Lisbonne : l’avenir d’un échec. Regards croisés sur l’économie, 1(11), 128–138. Cohen, E., Buigues, P.-A. (2014). Le décrochage industriel, Fayard, Paris. Cohen, E., Lorenzi, J.-H. (2000). Politiques industrielles pour l’Europe. La Documentation française, Paris. Commission des communautés européennes (1985). L’achèvement du marché intérieur. Livre blanc. Commission nationale d’évaluation des politiques d’innovation (2016). Quinze ans de politiques d’innovation en France. Rapport, France Stratégie, Paris. Condorcet, J.-A.-N. (1847a [1779]). Éloge de M. de Jussieu. In Œuvres de Condorcet, vol. 2, Condorcet O’Connor, A., Arago, M.F. (eds). Firmin Didot, Paris.

References

149

Condorcet, J.-A.-N. (1847b [1782]). Éloge de M. Duhamel. In Œuvres de Condorcet, vol. 2, Condorcet O’Connor, A., Arago, M.F. (eds). Firmin Didot, Paris. Condorcet, J.-A.-N. (1847c [1782b]). Éloge de M. de Vaucanson. Œuvres de Condorcet, vol. 3, Condorcet O’Connor, A., Arago, M.F. (eds). Firmin Didot, Paris. Condorcet, J.-A.-N. (1847d [1783]). Éloge de M. D’Alembert. In Œuvres de Condorcet, vol. 4, Condorcet O’Connor, A., Arago, M.F. (eds). Firmin Didot, Paris. Condorcet, J.-A.-N. (1847e [1786]). Vie de M. Turgot. In Œuvres de Condorcet, vol. 5, Condorcet O’Connor, A., Arago, M.F. (eds). Firmin Didot, Paris. Conférence nationale de l’industrie (2012). Ensemble, réindustrialiser la France pour la croissance et l’emploi. Report, CNI. Conseil d’analyse économique (2013). Évaluation des politiques publiques. Notes 1, CAE, Paris. Coquelin, C., Guillaumin, G.-U. (1852). Dictionnaire de l’économie politique. Guillaumin et Cie, Paris. Crozet, M. (2009). Commerce et géographie : la mondialisation selon Paul Krugman. Revue d’économie politique, 119(4), 513–534. Crozier, M. (1963). Le phénomène bureaucratique. Essai sur les tendances bureaucratiques des systèmes d’organisation modernes et sur leurs relations en France avec le système social et culturel. Le Seuil, Paris. DATAR (2003). Une nouvelle politique de développement des territoires pour la France. Report, Premier ministre, Paris. DATAR (2004). La France puissance industrielle : une nouvelle politique industrielle par les territoires. La Documentation française, Paris. Deaton, A. (2013). The Great Escape, Princeton University Press, Princeton. Defraigne, J.-C., Nouveau, P. (2017). Introduction à l’économie européenne, 2nd edition. De Boeck, Brussels. Del Rey, A. (2013). La tyrannie de l’évaluation. La Découverte, Paris. Delebarre, M., Garrigue, D. (2005). Le suivi de la stratégie de Lisbonne. Report, Assemblée nationale, Paris. Demmou, L. (2010). Le recul de l’emploi industriel en France entre 1980 et 2007. Ampleur et principaux déterminants : un état des lieux. Économie et Statistique, 438(1), 273–296.

150

Innovation and Industrial Policies

Dixit, A.K. (1996). The Making of Economic Policy: A Transaction Cost Politics Perspective. MIT Press, Cambridge, Massachusetts. Dixit, A.K., Stiglitz, J.-E. (1977). Monopolistic competition and optimum product diversity. American Economic Review, 67(3), 297–308. Domard, E. (1947). Expansion and employment. American Economic Review, 37(1), 34–55. Dosi, G., Freeman, C., Nelson, R., Silverberg, G., Soete, L. (1988). Technical Change and Economic Theory. Pinter Publishers, London. Dunoyer, C.-B. (1825). L’industrie et la morale considérées dans leurs rapports avec la liberté. Sautelet et Cie, Paris. Duranton, G., Puga, D. (2000). Diversity and specialization in cities: Why, where, and when does it matter?. Urban Studies, 37(3), 533–555. Duranton, G., Puga, D. (2001). Nursery cities: Urban diversity, process innovation, and the life cycle of products. American Economic Review, 91(5), 1454–1477. Duranton, G., Martin, P., Mayer, T., Mayneris, F. (2008). Les pôles de compétitivité, que peut-on en attendre ?. Rue d’Ulm, Paris. Encaoua, D. (2013). La notion de filière peut-elle être un outil de politique industrielle ?. In À quoi servent les filières ?, Bidet-Mayer, T., Toubal, L. (eds). Presses des Mines, Paris. Encaoua, D., Guesnerie, R. (2006). Politiques de la concurrence. La Documentation française, Paris. Enjolras, B. (2010). Gouvernance verticale, gouvernance horizontale et économie sociale et solidaire : le cas des services à la personne. Géographie, économie, société, 12(1), 15–30. États généraux de l’industrie (2010). Bilan de la concertation. Report, Ministère de l’Économie, de l’Industrie et de l’Emploi, Paris. Fitoussi, J.-P. (2013). Le théorème du lampadaire. Les liens qui libèrent, Paris. Flacher, D., Pelletan, J. (2007). Le concept d’industrie et sa mesure : origines, limites et perspectives. Une application à l’étude des mutations industrielles. Économie et statistique, 405/406, 13–46. Fogel, R. (1999). Catching up with the economy. American Economic Review, 89(1), 1–21. Fontagné, L., Lorenzi, J.-H. (2005). Désindustrialisation, délocalisations. La Documentation française, Paris.

References

151

Fontaine, P. (1992). Le concept d’industrie au XVIIIe siècle : à la recherche d’une acception oubliée. Économies et Sociétés, 26(3), 7–33. Fouquet, A. (2011). L’évaluation des politiques publiques en France, définition et historique. Revue française d'administration publique, 4(148), 9–16. Fouquet, A. (2013). L’évaluation des politiques publiques : état(s) de l’art et controverses. Revue française d’administration publique, 4(148), 835–847. Frison-Roche, M.-A. (2001). Le droit de la régulation. Le Dalloz, 7, 610–616. Furetière, A. (1690). Dictionnaire universel. Arnout et Reinier Leers, The Hague. Gaffard, J.-L. (1990). Économie industrielle et de l’innovation. Dalloz, Paris. Gaffard, J.-L. (2005). Vers une nouvelle politique industrielle. Lettre de l’OFCE, 269. Gaffard, J.-L. (2011). La croissance économique. Armand Colin, Paris. Gaffard, J.-L. (2013). Filières ou clusters : quel outil pour la politique industrielle ?. In À quoi servent les filières ?, Bidet-Mayer, T., Toubal, L. (eds). Presses des Mines, Paris. Gaffard, J.-L. (2017). L’industrie française entre déclin et renouveau. OFCE Policy Brief, 13, 1–14. Gaffard, J.-L., Krafft, J. (1996). Processus concurrentiel et politique communautaire en matière de concentration : introduction à une problématique. Revue de la Consommation et de la Concurrence, 89(1), 16–22. Gallois, L. (2012). Pacte pour la compétitivité de l’industrie française. Report, Commissariat général à l’investissement, Paris. Gallois, L. (2013). Préface. In À quoi servent les filières ?, Bidet-Mayer, T., Toubal, L. (eds). Presses des Mines, Paris. Georgescu-Roegen, N. (1970). The economics of production. American Economic Review, 60, 1–9. Givord, P. (2014). Méthodes économétriques pour l’évaluation de politiques publiques. Économie & Prévision, 204/205, 1–28. Godin, B. (2017). L’innovation sous tension : l’histoire d’un concept. Presses de l’université Laval, Quebec. Gouvernement (2013). Une nouvelle donne pour l’innovation. Press release, Ministère de l’Économie et des Finances, Paris. Gouvernement (2014). La nouvelle France industrielle. Présentation des feuilles de route des 34 plans de la nouvelle France industrielle. Report, CNI, Paris.

152

Innovation and Industrial Policies

Grossman, G., Helpman, E. (1990). Trade, innovation, and growth. American Economic Review, 80(2), 86–91. Grossman, G., Helpman, E. (1991). Innovation and Growth in the Global Economy. MIT Press, Cambridge, Massachusetts. Guellec, D. (2009). Économie de l’innovation. La Découverte, Paris. Guilhon, B. (2017). Innovation and Production Ecosystems. ISTE Ltd, London and John Wiley & Sons, New York. Hagg, O., Raulin, E., Souquet, C. (2004). Les relations interentreprises : nombreuses et d’abord hexagonales. Sessi, 195. Hannoun, M., Guerrier, G. (1996). Le partenariat industriel. Sessi, 183. Harfi, M., Lallement, R. (2016). Quinze ans de politiques d’innovation en France. Report, France Stratégie, Paris. Harfi, M., Lallement, R. (2017). Avis sur la politique des pôles de compétitivité. Report, France Stratégie, Paris. Harrisson, A., Rodriguez-Clare, A. (2010). Trade, foreign investment, and industrial policy. Handbook of Development Economics, 5, 4039–4214. Harrod, R. (1939). An essay in dynamic theory. Economic Journal, 49(193), 14–33. Harsin, P. (1930). De quand date le mot ‘industrie’ ?. Annales d’histoire économique et sociale, 2(6), 235–242. Hauser, H. (1925). Le mot ‘industrie’ chez Roland de la Platière. Revue historique, 150, 189–193. Hauser, H. (1927). Les débuts du capitalisme. Félix Alcan, Paris. Hayek, F. (1948 [1937]). Economic of knowledge. In Individualism and Economic Order, Hayek F. (ed.). University of Chicago Press, Chicago, 33–56. Hayek, F. (2002 [1968]). Competition as a discovery procedure. The Quarterly Journal of Austrian Economics, 5(3), 9–23. Helpman, E., Krugman, P. (1985). Market Structure and Foreign Trade: Increasing Returns, Imperfect Competition and The International Economy. Harvester Press, Brighton. Holcblat, N., Husson, M. (2013). L’industrie française. La Découverte, Paris. Jacobs, J. (1969). The Economy of Cities. Random House, New York. Jacquemin, A. (1979). Économie industrielle européenne, 2nd edition. Dunod, Paris.

References

153

Jaffe, A.-B., Henderson R., Trajtenberg M., (1993). Geographic localization of knowledge spillovers as evidenced by patents citations. Quarterly Journal of Economics, 108(3), 577–598. James, M. (1977). Pierre-Louis Roederer, Jean-Baptiste Say, and the concept of industrie. History of Political Economy, 9(4), 455–475. de Jaucourt, L. (1766a). Industrie. In Encyclopédie ou Dictionnaire raisonné des sciences, des arts et des métiers, vol. 8, Diderot, D., D’Alembert, J. (eds). Paris. de Jaucourt, L. (1766b). Innovation. In Encyclopédie ou Dictionnaire raisonné des sciences, des arts et des métiers, vol. 8, Diderot, D., D’Alembert, J. (eds). Paris. Jevakhoff, A., Cavailloles, D. (2017). Le rôle économique des fondations. Report, Inspection générale des finances, Paris. Juppé, A., Rocard, M. (2009). Investir pour l’avenir : priorités stratégiques d’investissement et emprunt national. La Documentation française, Paris. Kok, W. (2004). Relever le défi. La stratégie de Lisbonne pour la croissance et l’emploi. Office des publications officielles des communautés européennes, Luxembourg. Kremer, M. (1998). Patents buyouts: A mechanism for encouraging innovation. Quarterly Journal of Economics, 113(4), 1137–1167. Krugman, P.R. (1979). Increasing returns, monopolistic competition, and international trade. Journal of International Economics, 9(4), 469–479. Krugman, P.R. (1980). Scale economies, product differentiation, and the pattern of trade. American Economic Review, 70(5), 950–959. Krugman, P.R. (1984). Import protection as export promotion: International competition in the presence of oligopoly and economies of scale. In Monopolistic Competition and International Trade, Kierzkowski, H. (ed.). Oxford University Press, Oxford, 180–193. Krugman, P.R. (1987). Is free trade passé?. Journal of Economic Perspectives, 1(2), 131–144. Krugman, P.R. (1991). Increasing returns and economic geography. Journal of Political Economy, 99(3), 483–499. Krugman, P.R. (1993). The narrow and broad arguments for free trade. American Economic Review, 83(2), 362–366. Krugman, P.R. (1995). Development, Geography and Economic Theory. MIT Press, Cambridge, Massachusetts.

154

Innovation and Industrial Policies

Lacouette-Fougère, C., Lascoumes, P. (2013). L’évaluation : un marronnier de l’action gouvernementale ?. Revue française d’administration publique, 4(148), 859–875. Landes, D. (1988). The Unbound Prometheus. Cambridge University Press, Cambridge. Landier, A., Thesmar, D. (2013). Dix idées qui coulent la France. Flammarion, Paris. Laussel, D., Montet, C. (1989). La théorie du commerce internationale en concurrence imparfaite : développements récents et perspectives. In Commerce international en concurrence imparfaite, Laussel, D., Montet, C. (eds). Economica, Paris. Lauvergeon, A. (2013). Un principe et sept ambitions pour la France. Report, La Documentation française, Paris. Le Blanc, G. (2012). L’industrie aujourd’hui. Alternatives économiques, Special edition 93. Leijonhufvud, A. (1986). Capitalism and the factory system. In Economic as a Process: Essays in the New Institutional Economics, Langlois, R. (ed.). Cambridge University Press, Cambridge. Lévêque, F. (2004). Économie de la réglementation, 2nd edition. La Découverte, Paris. Levet, J.-L., Mathieu, C. (2013). Évaluation ex-post du Programme d’investissement d’avenir : un dispositif à construire. Report, La Documentation française, Paris. Lin, J.Y. (2013). La politique industrielle revisitée : une nouvelle perspective d’économie structurelle. Revue d’économie du développement, 21(2), 55–78. List, F. (1998 [1856]). The National System of Political Economy, English translation by G.A. Matile. J.B. Lippincott & Co., Philadelphia. Loasby, B. (1998). The concept of capabilities. In Economic Organization, Capabilities and Co-ordination, Essays in Honour of G.B. Richardson, Foss, N., Loasby, B. (eds). Routledge, London, 163–182. Lombard, D. (2017). Nouvelle économie, nouvelle industrie. Odile Jacob, Paris. Lucas, R.-E. (1988). On the mechanics of economic development. Journal of Monetary Economics, 22, 3–42. Maddison, A. (2001). L’économie mondiale : une perspective millénaire. OCDE, Paris. Maillet, P. (1984). La politique industrielle. PUF, Paris.

References

155

Mansouri-Guilani, N. (2015). Promouvoir une culture de l’évaluation des politiques publiques. Conseil Économique Social et Environnemental [Online]. Available: https://www.lecese.fr/travaux-publies/promouvoir-une-culture-de-l-valuation-despolitiques-publiques. Marshall, A. (1979 [1890]). Principles of Economics, 8th edition. Macmillan, London. Marty, F. (2004). Analyse de la libéralisation des industries de réseaux au travers de la jurisprudence du Conseil de la concurrence français. Revue internationale de droit économique, 18(4), 435–459. Maystadt, P. (2016). Programme d’investissement d’avenir, rapport du comité d’évaluation à mi-parcours. Report, France Stratégie, Paris. McNulty, P.-J. (1968). Economic theory and the meaning of competition. Quarterly Journal of Economics, 82(4), 639–656. Ménard, C. (1993). L’économie des organisations. La Découverte, Paris. Migaud, D. (2013). Les cinq défis de l’évaluation. Revue française d’administration publique, 4(148), 849–858. Morvan, Y. (1991). Fondements d’économie industrielle, 2nd edition. Economica, Gestion, Paris. Mounier-Kuhn, P.-E. (1994). Le Plan Calcul, Bull et l’industrie des composants : les contradictions d’une stratégie. Revue historique, 591, 123–153. Muller, P. (2004). L’État en action revisité. Pôle Sud, 21, 31–42. Musgrave, R. (1959). A Theory of Public Finance. McGraw Hill, New York. National Bureau of Economic Research (1962). The Rate and Direction of Inventive Activity: Economic and Social Factors. Princeton University Press, Princeton. Nelson, R. (1959). The simple economics of basic scientific research. Journal of Political Economy, 67(3), 297–306. Nelson, R., Winter, S. (1982). An Evolutionary Theory of Economic Change. Harvard University Press, Cambridge. Penrose, E. (1995 [1959]). The Theory of the Growth of the Firm, 3rd edition. Oxford University Press, Oxford. Perrot, A. (2002). Les frontières entre régulation sectorielle et politique de la concurrence. Revue française d’économie, 16(4), 81–112. Picart, C. (2006). Les gazelles en France. In Une stratégie PME pour la France, Betbèze, J.-P., Saint-Etienne, C. (eds). La Documentation française, Paris, 77–124.

156

Innovation and Industrial Policies

Premier ministre (2012). Pacte national pour la croissance, la compétitivité et l’emploi. Press release, Matignon, Paris. Quéré, M., Ravix, J.-L. (1995). Le chercheur-entrepreneur dans la dynamique des relations recherche-industrie : un cadre d’analyse. In Économie de la connaissance et organisations, Guilhon, B., Huard, P., Orillard, M., Zimmerman, J.-B. (eds). L’Harmattan, Paris, 208–229. Quéré, M., Ravix, J.-L., Ravix, J.-T., Romani, P.M. (1997). Frontières de la firme, division institutionnelle du travail et processus de concurrence. In Les frontières de la firme, Garrouste, P. (ed.). Economica, Paris, 105–133. Ravix, J.-L. (1990). L’émergence de la firme et des coopérations inter-firmes dans la théorie de l’organisation industrielle : Coase et Richardson. Revue d’économie industrielle, 51(1), 27–50. Ravix, J.-T. (1997). L’économie des liaisons industrielles. In Séminaire sur les nouvelles formes de relation d’emploi et le contrat d’activité. Commissariat Général du Plan, Paris. Ravix, J.-L. (1998). Co-operation and competition paradoxes in the theory of the organisation of industry. In Economic Organization, Capabilities and Coordination: Essays in Honor of G.B. Richardson, Foss, N., Loasby, B. (eds). Routledge, London, 63–83. Ravix, J.-T. (2001). Co-operation and the organization of the industry. In The Dynamics of Industrial Collaboration, Bellon, B., Plunket, A., Voisin, C. (eds). Edward Elgar, London, 246–259. Ravix, J.-T. (2005). Relations interentreprises et frontières des industries. INSEE Méthodes, 109. Ravix, J.-T. (2013). Le programme d’investissement d’avenir : une nouvelle étape dans l’évolution de la notion de politique industrielle. In Annexe 4, Levet, J.-L., Mathieu C. (eds), 99–104. Richardson, G.-B. (1972). The organisation of industry. The Economic Journal, 82(327), 883–896. Richardson, G.-B. (1975). Adam Smith on competition and increasing returns. In Essays on Adam Smith, Skinner, A.S., Wilson, T. (eds). Clarendon Press, Oxford, 350–360. Richardson, G.-B. (1990 [1960]). Information and Investment, 2nd edition. Clarendon Press, Oxford. Robinson, J. (2010). Politique industrielle et développement : analyse en termes d’économie politique. Revue d’économie du développement, 18(4), 21–45.

References

157

Rodrik, D. (2008). Industrial policy: Don’t ask why, ask how. Middle East Development Journal, 1(1), 1–29. Rodrik, D. (2010). Le retour de la politique industrielle. Projet Syndicate [Online]. Available: https://www.project-syndicate.org/commentary/the-return-of-industrialpolicy/french?barrier=accesspaylog. Rodrik, D. (2011). L’impératif du secteur manufacturier. Projet Syndicate, 10 August [Online]. Available at: https://www.project-syndicate.org/commentary/ the-manufacturing-imperative/french?barrier=accesspaylog. Roland de la Platière, J.-M. (1790). Discours préliminaire. Encyclopédie méthodique. Manufactures, arts et métiers, vol. 1. Panckoucke, Paris. Romer, P.-M. (1986). Increasing returns and long-run growth. Journal of Political Economy, 94(5), 1002–1037. Romer, P.-M. (1990). Endogenous technological change. Journal of Political Economy, 98(5), 71–102. Romer, P.-M. (1993). Implementing a national technology strategy with selforganizing industry investment boards. Brookings Paper on Economic Activity, 24(2), 345–399. Saint-Paul, G. (1996). Les nouvelles théories de la croissance et leurs implications pour la politique économique et l’analyse de la concurrence internationale. Revue française d’économie, 11(3), 3–20. Saviotti, P.P. (1995). Renouveau des politiques industrielles : le point de vue des théories évolutionnistes. Revue d’économie industrielle, 71(1), 199–212. Saviotti, P.P., Metcalfe, J.S. (eds) (1991). Evolutionary Theories of Economic and Technological Change. Routledge, London. Say, J.-B. (1840). Cours complet d’économie politique pratique, 2nd edition. Guillaumin et Cie, Paris. Say, J.-B. (1841). Traité d’économie politique, 6th edition. Guillaumin et Cie, Paris. Say, L., Chaillet, J. (1900). Nouveau dictionnaire d’économie politique, 2nd edition. Guillaumin et Cie, Paris. Scandella, J.-S. (2008). Les pôles, réseaux d’excellence et d’innovation. Autrement, Paris. Schawb, K. (2016). The Fourth Industrial Revolution. World Economic Forum, Geneva. Schumpeter, J.A. (1934 [1911]). The Theory of Economic Development. Harvard University Press, Cambridge.

158

Innovation and Industrial Policies

Schumpeter, J.A. (1983). Histoire de l’analyse économique, Gallimard, Paris. Schumpeter, J.A. (1990 [1942]). Capitalisme, socialisme et démocratie. Payot, Paris. Sée, H. (1925). À propos du mot ‘industrie’. Revue Historique, 159, 58–61. Sée, H. (1928). Histoire économique et sociale (1927–1928). Revue Historique, 158, 297–335. Smith, A. (1991 [1776]). Recherches sur la nature et les causes de la richesse des nations. Flammarion, Paris. Solow, R. (1956). A contribution to the theory of economic growth. Quarterly Journal of Economics, 70(1), 65–94. Solow, R. (1957). Technical change and the aggregate production function. Review of Economics and Statistics, 39(3), 312–320. Spencer, B., Brander, J. (1983). International R&D rivalry and industrial strategy. Review of Economic Studies, 50(4), 707–722. Sraffa, P. (1926). The laws of return under competitive conditions. The Economic Journal, 36(144), 535–550. Sraffa, P. (1975). Écrits d’économie politique. Economica, Paris. Stiglitz, J.-E., Greenwald, B.C. (2014). Creating a Learning Society. Columbia University Press, New York. Stiglitz, J.-E., Lin J.Y., Monga, C. (2013). Introduction: The Rejuvenation of Industrial Policy. In The Industrial Policy Revolution I. The Role of Government Beyond Ideology, Stiglitz, J.-E., Lin J.Y. (eds). Palgrave Macmillan, London, 2–15. Stoffaes, C. (1980). Politique industrielle et filières. Revue d’économie industrielle, 13(3), 86–99. Stoffaes, C. (2011). Le rôle du Corps des Mines dans la politique industrielle française : deux siècles d’action et d’influence. Annales des Mines, 4, 48–67. Stoleru, L. (1989). L’évaluation : une indispensable autodiscipline. In L’évaluation des politiques et des actions publiques. Propositions en vue de l’évaluation du revenu minimum d’insertion, Viveret, P. (ed.). La Documentation française, Paris, 5–6. Tavernier, J.-L. (2011). Investissements d’avenir et pôles de compétitivité. Brief, Commissariat général à l’Investissement, Paris. Thibault, G. (2008). Quelle stratégie industrielle pour la France face à la mondialisation ?. Technip, Paris.

References

159

Tirole, J. (2016). Économie du bien commun. PUF, Paris. Vandamme, J., Simons, E. (1990). Le contrôle des concentrations dans la communauté européenne. Courrier hebdomadaire, 28(1293), 1–38. Viveret, P. (1989). L’évaluation des politiques et actions publiques. Report, La Documentation française, Paris. Warlouzet, L. (2008). La France, la construction européenne et les limites de la conversion à la ‘libre concurrence’. Parlement(s) Revue d’histoire politique, Special edition 3, 25–37. Warlouzet, L. (2012). La contribution européenne aux projets de régulation mondiale de la concurrence (1945–2005). Les cahiers IRICE, 9, 105–114. Weil, T., Fen Chon, S. (2008). Les pôles de compétitivité français. Futuribles, 342, 5–26. Williamson, O.E. (1968). Economies as an antitrust defense: The welfare tradeoff. American Economic Review, 58(1), 18–36. Williamson, O.E. (1975). Markets and Hierarchy. Free Press, New York. Williamson, O.E. (1985). The Economic Institutions of Capitalism: Firms, Markets, Relational Contracting. Free Press, New York. Williamson J. (1990). What Washington means by policy reform?. In Latin American Adjustment: How Much Has Happened?. Williamson, J. (ed.). Institute for International Economics, Washington. World Bank (2008). The Growth Report, Strategies for Sustained Growth and Inclusive Development. Commission on Growth and Development, Washington. Young, A.A. (1928). Increasing returns and economic progress. The Economic Journal, 38. Zarka, Y.-C. (2009). L’évaluation : un pouvoir supposé savoir. Cités, 1(37), 113–123.

Index

A, B

D, E

agglomeration, 38, 62, 63, 82, 84, 85, 123 authority independent administrative, 20 sectoral regulatory, 30–32 balance, 9, 13, 31, 54–57, 59, 63, 124, 130, 142

deindustrialization, 2–6, 15, 64, 95 102 deregulation, 21, 27, 29 economy market, 10, 14 geography, 55, 64 ecosystem, 5, 65, 71, 75, 106, 107 education, 65, 67, 123 entrepreneur, 117 entreprise, 12, 14, 25, 26, 28, 33, 35, 37, 42, 45, 49, 75, 77, 79, 80, 85, 102, 106, 111, 112, 115–119, 139 public, 26, 28, 85 European Commission, 15, 22–28, 33–36, 43–45, 66, 68 European Union, 21, 23, 28, 50, 55, 63, 65, 68, 75 evaluation, 6, 35, 49, 66, 72, 82, 86, 89, 90, 93, 94, 97, 108, 124, 133–141 externality, 55, 58

C cluster, 65, 84, 94, 106 competition, 1, 4, 6, 7, 9, 10, 12–14, 19, 20, 22, 23, 25–28, 31–35, 37, 38, 41, 43, 46, 49–51, 53–55, 58–63, 68, 77, 83, 85, 88, 102, 120, 121, 127–131, 141, 142 imperfect, 55, 60, 61, 131 monopolistic, 54, 55, 62, 63, 121 perfect, 9, 38, 60, 61, 63, 130 competitive order, 21, 25, 26, 38 competitiveness, 68, 70, 75, 76, 83, 94, 100–102 clusters, 70, 71, 73, 77, 82, 83, 93, 94, 98, 107, 140 concentration, 14, 32, 34–37, 73, 83, 84, 121, 128 crisis, 9, 13, 34, 45, 127

F, G free trade, 22, 61, 63, 127 freedom of movement, 21–23, 48 funding, 13, 25, 27, 71, 76, 77, 79, 86, 89, 90, 92, 97, 102, 129, 139, 140

Innovation and Industrial Policies, First Edition. Joël-Thomas Ravix and Marc Deschamps. © ISTE Ltd 2019. Published by ISTE Ltd and John Wiley & Sons, Inc.

162

Innovation and Industrial Policies

globalization, 98 governance, 72, 82, 85–87, 90–95, 99, 104, 128 government, 12, 32, 40, 46, 59, 72, 77, 81, 92, 99, 101–103, 123, 125 growth, 2, 6, 9, 12, 29, 51, 53–60, 64, 65, 67–69, 71, 78, 79, 81, 85, 88, 97, 100–103, 124, 126, 128, 129, 138 economic, 2, 53, 55, 59, 65, 66, 69, 97 endogenous, 51, 57–59, 64, 65 I increasing returns to scale, 54, 55, 60, 63 industry, 2–5, 7–11, 13, 15, 19, 37, 39, 61, 68–70, 72, 77, 81, 87, 94, 97–102, 104, 107, 109–116, 119–123, 125, 127, 131, 135, 140, 142 network, 21, 27–29, 31 industrial organization, 110, 111, 113, 115 Industrial Revolution, 1, 51, 102, 122 infrastructure, 28, 31 innovation, 4, 5, 51, 54, 55, 58–60, 62, 64, 65, 68–72, 74–78, 80–84, 87, 89, 91, 95, 97, 99, 100, 102, 103, 105–107, 122, 126–131, 135, 137–140, 142 international trade, 53, 55, 60, 62, 63 investment, 10, 12, 14, 57, 79, 80, 86–94, 97, 101, 117, 125, 137, 138 J, K, L jurisprudential construction, 25, 28, 42 knowledge, 8, 16, 51, 56–58, 64, 65, 67, 71, 88, 115, 116, 120–122, 126, 127, 139

law competition, 21, 25, 31, 35, 38 European, 20, 22–25, 35, 38, 42, 46, 47 liberalization, 27–29, 31, 53, 68, 130 local authorities, 82, 137 local productive systems, 75, 83 M, N, O market, 4, 21, 23, 32, 36, 45, 49, 63, 86, 100, 105, 111, 112, 114, 118–120, 124, 128 failure, 6, 16 public, 46 monopoly, 26, 28, 29, 31 natural, 28 national champions, 13–15, 18, 32, 82 neoliberal, 18, 54 new economic geography, 60, 63 industrial policy, 54, 69–71, 75, 76, 81, 82, 85, 87, 88 oligopoly, 60 P, R Pareto optimum, 38 patent, 25 policy, 6–22, 26, 27, 29, 32, 34, 38, 41, 45, 46, 48, 50, 51, 53–56, 60, 61, 63, 65, 68–70, 72, 73, 76, 79, 82–86, 92–95, 97–107, 122–130, 133–136, 138–142 competition, 6, 7, 12, 13, 16, 19–22, 34, 38, 41, 51, 65, 106, 107, 122, 129 industrial, 1, 6–16, 18–20, 27, 32, 38, 41, 45, 46, 50, 51, 68, 69, 76, 82, 83, 85, 86, 92–95, 97–99, 102–107, 122–126, 128–130, 140

Index

innovation, 20, 53, 54, 69, 70, 94, 97, 99, 102, 103, 107, 122, 129, 130, 138 microeconomic, 29 strategy trade, 60, 61 product differentiation, 10, 55, 60 productive apparatus, 3, 9, 11, 32, 38 productivity, 125, 139 program Future Investment (programme d’investissement d’avenir), 54, 87, 91, 97, 138, 140 industrial, 77 public action, 19, 86, 87, 126, 133, 134, 136 authorities, 11, 20, 40, 51, 53, 54, 69, 75, 79, 83, 86, 98, 99, 101, 102, 104, 127, 137 good, 59 procurement, 19, 38, 45–50, 75, 85 service, 25, 42, 134

163

regulation, 19, 20, 27, 29–31, 125 competitive, 20 relocation, 4 S, T sector, 19, 99–101, 104–107 SME, 12, 14, 15, 68, 69, 73, 75–81, 85, 87, 89, 91, 92, 94, 102, 103 innovative, 69, 78, 89 state, 110, 117, 121, 124, 130, 141, 142 State aid, 42, 43, 45, 123 State subsidies, 38 sustainable development, 66, 67, 87, 88, 91, 94, 139 technical progress, 35, 56–58 territory, 12, 13, 24, 27, 35, 40, 53, 70–72, 74, 75, 83, 84, 94, 106 transport cost, 63, 98

Other titles from

in Innovation, Entrepreneurship and Management

2019 AMENDOLA Mario, GAFFARD Jean-Luc Disorder and Public Concern Around Globalization BARBAROUX Pierre Disruptive Technology and Defence Innovation Ecosystems (Innovation in Engineering and Technology Set – Volume 5) DOU Henri, JUILLET Alain, CLERC Philippe Strategic Intelligence for the Future 1: A New Strategic and Operational Approach Strategic Intelligence for the Future 2: A New Information Function Approach FRIKHA Azza Measurement in Marketing: Operationalization of Latent Constructs FRIMOUSSE Soufyane Innovation and Agility in the Digital Age (Human Resources Management Set – Volume 2)

GAY Claudine, SZOSTAK Bérangère L. Innovation and Creativity in SMEs: Challenges, Evolutions and Prospects (Smart Innovation Set – Volume 21) GORIA Stéphane, HUMBERT Pierre, ROUSSEL Benoît Information, Knowledge and Agile Creativity (Smart Innovation Set – Volume 22) HELLER David, DE CHADIRAC Sylvain, HALAOUI Lana, JOUVET Camille The Emergence of Start-ups (Economic Growth Set – Volume 1) HÉRAUD Jean-Alain, KERR Fiona, BURGER-HELMCHEN Thierry Creative Management of Complex Systems (Smart Innovation Set – Volume 19) LATOUCHE Pascal Open Innovation: Corporate Incubator (Innovation and Technology Set – Volume 7) LEHMANN Paul-Jacques The Future of the Euro Currency LEIGNEL Jean-Louis, MÉNAGER Emmanuel, YABLONSKY Serge Sustainable Enterprise Performance: A Comprehensive Evaluation Method LIÈVRE Pascal, AUBRY Monique, GAREL Gilles Management of Extreme Situations: From Polar Expeditions to ExplorationOriented Organizations MILLOT Michel Embarrassment of Product Choices 2: Towards a Society of Well-being N’GOALA Gilles, PEZ-PÉRARD Virginie, PRIM-ALLAZ Isabelle Augmented Customer Strategy: CRM in the Digital Age NIKOLOVA Blagovesta The RRI Challenge: Responsibilization in a State of Tension with Market Regulation (Innovation and Responsibility Set – Volume 3)

PRIOLON Joël Financial Markets for Commodities QUINIOU Matthieu Blockchain: The Advent of Disintermediation ROGER Alain, VINOT Didier Skills Management: New Applications, New Questions (Human Resources Management Set – Volume 1) SERVAJEAN-HILST Romaric Co-innovation Dynamics: The Management of Client-Supplier Interactions for Open Innovation (Smart Innovation Set – Volume 20) SKIADAS Christos H., BOZEMAN James R. Data Analysis and Applications 1: Clustering and Regression, Modelingestimating, Forecasting and Data Mining (Big Data, Artificial Intelligence and Data Analysis Set – Volume 2) Data Analysis and Applications 2: Utilization of Results in Europe and Other Topics (Big Data, Artificial Intelligence and Data Analysis Set – Volume 3)

2018 BURKHARDT Kirsten Private Equity Firms: Their Role in the Formation of Strategic Alliances CALLENS Stéphane Creative Globalization (Smart Innovation Set – Volume 16) CASADELLA Vanessa Innovation Systems in Emerging Economies: MINT – Mexico, Indonesia, Nigeria, Turkey (Smart Innovation Set – Volume 18) CHOUTEAU Marianne, FOREST Joëlle, NGUYEN Céline Science, Technology and Innovation Culture (Innovation in Engineering and Technology Set – Volume 3)

CORLOSQUET-HABART Marine, JANSSEN Jacques Big Data for Insurance Companies (Big Data, Artificial Intelligence and Data Analysis Set – Volume 1) CROS Françoise Innovation and Society (Smart Innovation Set – Volume 15) DEBREF Romain Environmental Innovation and Ecodesign: Certainties and Controversies (Smart Innovation Set – Volume 17) DOMINGUEZ Noémie SME Internationalization Strategies: Innovation to Conquer New Markets ERMINE Jean-Louis Knowledge Management: The Creative Loop (Innovation and Technology Set – Volume 5) GILBERT Patrick, BOBADILLA Natalia, GASTALDI Lise, LE BOULAIRE Martine, LELEBINA Olga Innovation, Research and Development Management IBRAHIMI Mohammed Mergers & Acquisitions: Theory, Strategy, Finance LEMAÎTRE Denis Training Engineers for Innovation LÉVY Aldo, BEN BOUHENI Faten, AMMI Chantal Financial Management: USGAAP and IFRS Standards (Innovation and Technology Set – Volume 6) MILLOT Michel Embarrassment of Product Choices 1: How to Consume Differently PANSERA Mario, OWEN Richard Innovation and Development: The Politics at the Bottom of the Pyramid (Innovation and Responsibility Set – Volume 2) RICHEZ Yves Corporate Talent Detection and Development

SACHETTI Philippe, ZUPPINGER Thibaud New Technologies and Branding (Innovation and Technology Set – Volume 4) SAMIER Henri Intuition, Creativity, Innovation TEMPLE Ludovic, COMPAORÉ SAWADOGO Eveline M.F.W. Innovation Processes in Agro-Ecological Transitions in Developing Countries (Innovation in Engineering and Technology Set – Volume 2) UZUNIDIS Dimitri Collective Innovation Processes: Principles and Practices (Innovation in Engineering and Technology Set – Volume 4) VAN HOOREBEKE Delphine The Management of Living Beings or Emo-management

2017 AÏT-EL-HADJ Smaïl The Ongoing Technological System (Smart Innovation Set – Volume 11) BAUDRY Marc, DUMONT Béatrice Patents: Prompting or Restricting Innovation? (Smart Innovation Set – Volume 12) BÉRARD Céline, TEYSSIER Christine Risk Management: Lever for SME Development and Stakeholder Value Creation CHALENÇON Ludivine Location Strategies and Value Creation of International Mergers and Acquisitions CHAUVEL Danièle, BORZILLO Stefano The Innovative Company: An Ill-defined Object (Innovation Between Risk and Reward Set – Volume 1)

CORSI Patrick Going Past Limits To Growth D’ANDRIA Aude, GABARRET

Inés Building 21st Century Entrepreneurship (Innovation and Technology Set – Volume 2) DAIDJ Nabyla Cooperation, Coopetition and Innovation (Innovation and Technology Set – Volume 3) FERNEZ-WALCH Sandrine The Multiple Facets of Innovation Project Management (Innovation between Risk and Reward Set – Volume 4) FOREST Joëlle Creative Rationality and Innovation (Smart Innovation Set – Volume 14) GUILHON Bernard Innovation and Production Ecosystems (Innovation between Risk and Reward Set – Volume 2) HAMMOUDI Abdelhakim, DAIDJ Nabyla Game Theory Approach to Managerial Strategies and Value Creation (Diverse and Global Perspectives on Value Creation Set – Volume 3) LALLEMENT Rémi Intellectual Property and Innovation Protection: New Practices and New Policy Issues (Innovation between Risk and Reward Set – Volume 3) LAPERCHE Blandine Enterprise Knowledge Capital (Smart Innovation Set – Volume 13) LEBERT Didier, EL YOUNSI Hafida International Specialization Dynamics (Smart Innovation Set – Volume 9)

MAESSCHALCK Marc Reflexive Governance for Research and Innovative Knowledge (Responsible Research and Innovation Set – Volume 6) MASSOTTE Pierre Ethics in Social Networking and Business 1: Theory, Practice and Current Recommendations Ethics in Social Networking and Business 2: The Future and Changing Paradigms MASSOTTE Pierre, CORSI Patrick Smart Decisions in Complex Systems MEDINA Mercedes, HERRERO Mónica, URGELLÉS Alicia Current and Emerging Issues in the Audiovisual Industry (Diverse and Global Perspectives on Value Creation Set – Volume 1) MICHAUD Thomas Innovation, Between Science and Science Fiction (Smart Innovation Set – Volume 10) PELLÉ Sophie Business, Innovation and Responsibility (Responsible Research and Innovation Set – Volume 7) SAVIGNAC Emmanuelle The Gamification of Work: The Use of Games in the Workplace SUGAHARA Satoshi, DAIDJ Nabyla, USHIO Sumitaka Value Creation in Management Accounting and Strategic Management: An Integrated Approach (Diverse and Global Perspectives on Value Creation Set –Volume 2) UZUNIDIS Dimitri, SAULAIS Pierre Innovation Engines: Entrepreneurs and Enterprises in a Turbulent World (Innovation in Engineering and Technology Set – Volume 1)

2016 BARBAROUX Pierre, ATTOUR Amel, SCHENK Eric Knowledge Management and Innovation (Smart Innovation Set – Volume 6) BEN BOUHENI Faten, AMMI Chantal, LEVY Aldo Banking Governance, Performance And Risk-Taking: Conventional Banks Vs Islamic Banks BOUTILLIER Sophie, CARRÉ Denis, LEVRATTO Nadine Entrepreneurial Ecosystems (Smart Innovation Set – Volume 2) BOUTILLIER Sophie, UZUNIDIS Dimitri The Entrepreneur (Smart Innovation Set – Volume 8) BOUVARD Patricia, SUZANNE Hervé Collective Intelligence Development in Business GALLAUD Delphine, LAPERCHE Blandine Circular Economy, Industrial Ecology and Short Supply Chains (Smart Innovation Set – Volume 4) GUERRIER Claudine Security and Privacy in the Digital Era (Innovation and Technology Set – Volume 1) MEGHOUAR Hicham Corporate Takeover Targets MONINO Jean-Louis, SEDKAOUI Soraya Big Data, Open Data and Data Development (Smart Innovation Set – Volume 3) MOREL Laure, LE ROUX Serge Fab Labs: Innovative User (Smart Innovation Set – Volume 5) PICARD Fabienne, TANGUY Corinne Innovations and Techno-ecological Transition (Smart Innovation Set – Volume 7)

2015 CASADELLA Vanessa, LIU Zeting, DIMITRI Uzunidis Innovation Capabilities and Economic Development in Open Economies (Smart Innovation Set – Volume 1) CORSI Patrick, MORIN Dominique Sequencing Apple’s DNA CORSI Patrick, NEAU Erwan Innovation Capability Maturity Model FAIVRE-TAVIGNOT Bénédicte Social Business and Base of the Pyramid GODÉ Cécile Team Coordination in Extreme Environments MAILLARD Pierre Competitive Quality and Innovation MASSOTTE Pierre, CORSI Patrick Operationalizing Sustainability MASSOTTE Pierre, CORSI Patrick Sustainability Calling

2014 DUBÉ Jean, LEGROS Diègo Spatial Econometrics Using Microdata LESCA Humbert, LESCA Nicolas Strategic Decisions and Weak Signals

2013 HABART-CORLOSQUET Marine, JANSSEN Jacques, MANCA Raimondo VaR Methodology for Non-Gaussian Finance

2012 DAL PONT Jean-Pierre Process Engineering and Industrial Management MAILLARD Pierre Competitive Quality Strategies POMEROL Jean-Charles Decision-Making and Action SZYLAR Christian UCITS Handbook

2011 LESCA Nicolas Environmental Scanning and Sustainable Development LESCA Nicolas, LESCA Humbert Weak Signals for Strategic Intelligence: Anticipation Tool for Managers MERCIER-LAURENT Eunika Innovation Ecosystems

2010 SZYLAR Christian Risk Management under UCITS III/IV

2009 COHEN Corine Business Intelligence ZANINETTI Jean-Marc Sustainable Development in the USA

2008 CORSI Patrick, DULIEU Mike The Marketing of Technology Intensive Products and Services DZEVER Sam, JAUSSAUD Jacques, ANDREOSSO Bernadette Evolving Corporate Structures and Cultures in Asia: Impact of Globalization

2007 AMMI Chantal Global Consumer Behavior

2006 BOUGHZALA Imed, ERMINE Jean-Louis Trends in Enterprise Knowledge Management CORSI Patrick et al. Innovation Engineering: the Power of Intangible Networks

WILEY END USER LICENSE AGREEMENT Go to www.wiley.com/go/eula to access Wiley’s ebook EULA.