Routledge Handbook of Evolutionary Economics [1 ed.] 9780367025687, 9781032533391, 9780429398971

Table of contents :
Cover
Half Title
Title Page
Copyright Page
Contents
Contributors
Evolutionary economics: A navigational guide
Note
Part I: Foundational issues and theoretical domains
1. Joseph A. Schumpeter: One of the founders of evolutionary economics
1.1. Introduction
1.2. Charles Darwin and biological evolution
1.3. Schumpeter on Darwin and Darwinian ideas in economics
1.3.1. Schumpeter's "Monroe doctrine"
1.3.2. Evolution - A directionless process
1.4. Schumpeter on Engels and Marx on Darwin
1.4.1. Darwinism - The natural science foundation of class struggle?
1.4.2. A discovery as important as that of the heliocentric system
1.5. Making his sense of Marx
1.5.1. Marx interpreted in a "conservative sense"
1.5.2. "Law" of the falling rate of profit
1.5.3. The "capitalist engine"
1.6. Ideas have consequences
1.6.1. Ideas vs. facts
1.6.2. The "spirit" of capitalism
1.6.3. A "culture" of innovation
1.6.4. Schumpeter vs. Marx
1.7. Social conflicts - A flywheel of evolution
1.7.1. Confounding capitalists and entrepreneurs
1.7.2. The dynamic performance of capitalism and socialism
1.7.3. Income distribution and social justice
1.7.4. Stabilisation policy
1.7.5. Income distribution policy
1.8. Culture and democracy
1.8.1. Culture
1.8.2. Democracy
1.9. Concluding observations
Notes
References
2. Thorstein Bunde Veblen: A founder of evolutionary economics
2.1. Introduction
2.2. Theoretical approach and evolutionary epistemology
2.3. Applied evolutionary economics: The theory of the leisure class
2.4. Evolutionary heterodox supply-side economics: Technostructure, absentee owners, and speculative finance
2.5. Veblen as a precursor of evolutionary economics and his current significance
References
3. The foundational evolutionary traverse of Richard R. Nelson and Sidney G. Winter
3.1. The context
3.2. Economic order and firm theory in the early Sidney Winter
3.3. The problem of economic change in the early Richard Nelson
3.4. The foundational evolutionary traverse
3.5. The golden decades
3.6. Far beyond the consolidated beachhead
Acknowledgments
References
4. F.A. Hayek and evolutionary Austrian economics
4.1. Carl Menger and his heirs
4.2. Carl Menger's evolutionary social theory
4.3. F.A. Hayek's Darwinian theory of cultural evolution
4.4. Ludwig von Mises' praxiological, aprioristic subjectivism
4.5. F.A. Hayeks's evolutionary, naturalistic subjectivism
4.6. Conclusion
Notes
References
5. Kenneth Boulding's contribution to evolutionary economics
5.1. Introduction
5.2. What is Boulding's relevance as a founding father of evolutionary economics?
5.3. What evolves in Boulding's evolutionary economics?
5.4. The key influences on and elements of Boulding's evolutionary economics
5.5. Theory of production
5.6. Theory of the market
5.7. Government intervention
Notes
References
6. Evolutionary economics and psychology: Where we are, where we could go
6.1. Introduction: A good foundation for further building
6.2. Existing foundations: Neo-Schumpeterian, Veblenian and naturalistic perspectives
6.3. Neo-Schumpeterian perspectives: Cognitive psychology
6.4. Veblenian perspectives: Habit psychology
6.5. Naturalistic and bioeconomic perspectives: Evolutionary psychology
6.6. Opportunities for growth: Personality, social, affective, and persuasion psychology
6.7. Personality psychology
6.8. Social psychology
6.9. Affective psychology
6.10. Persuasion psychology
6.11. Conclusion: Where we are, where we could go
References
7. Evolutionary cultural science
7.1. Introduction
7.2. Ontology and evolutionary methodology
7.3. Principles of ECS
7.4. ECS behavioural theory of consumption
7.5. Outlook: The culture of economics
References
8. Evolutionary economics and economic history
8.1. Ontological and heuristic foundations of evolutionary economics and mainstream economics
8.2. History of history
8.3. Mutual considerations: Evolutionary economics as economic history
Notes
References
9. Why an evolutionary economic geography?: The spatial economy as a complex evolving system
9.1. Introduction: The geographical foundations of the economy
9.2. The scope and contribution of evolutionary economic geography
9.2.1. Three key guiding principles
9.2.2. Putting the principles in place
9.3. The evolution of evolutionary economic geography: Future challenges and directions
9.3.1. Widening the field of focus
9.3.2. Moving beyond 'patchwork' evolutionary economic geography: In search of a unifying theoretical framework?
9.3.3. The importance of addressing key spatial policy issues
9.4. Conclusion
Notes
References
10. Darwin's ideas and their mixed reception in evolutionary economics
10.1. Introduction
10.2. Four positions regarding the relevance of Darwin's ideas for evolutionary economics
10.2.1. Metaphors and analogies
10.2.2. Generalized darwinism
10.2.3. The continuity hypothesis
10.2.4. Complexity theory as an improvement of Darwin's ideas
10.3. The heuristics and epistemological principles implied in each position
10.4. Conclusions
Notes
References
11. Computational evolutionary economics: Minimal principle and minimum intelligence
11.1. Backgrounds and motivation
11.1.1. Minimum principle
11.1.2. Hierarchies
11.2. Simplicity principle
11.2.1. Minimum description length
11.2.2. Is the SP the MinP?
11.2.3. Maximum entropy principle
11.3. Minimum intelligence in ACE
11.3.1. Minimum intelligence by principles
11.3.2. Minimum intelligence in practice
11.4. Neural nets and minimality
11.5. Concluding remarks
Acknowledgments
Notes
References
12. Evolutionary modeling and the rule-based approach
12.1. What else if not optimization?
12.2. Robinson Crusoe as a rule-user and -maker
12.2.1. Robinson in isolation
12.2.2. Institutions and social object rules
12.2.3. Object rules: Technical change
12.3. Economic evolution
12.4. Generic rules and trajectories
12.5. Empirical quest
12.6. Benefits of the taxonomy
12.7. Conclusion
12.8. Classification of rules
12.9. Biography
Notes
References
13. Contingency in evolutionary economics: Causality and comparative analysis Marco Lehmann-Waffenschmidt
13.1. Comparative evolutionary analysis
13.2. The graphical-analytical contingency concept
13.3. The gradual measurement of causality relationships between different states of a process by the degree of prograde causality and the degree of retrograde causality
13.4. On the relationship between the contingency approach and the causal-logical terms "necessary" and "sufficient"
13.5. The extension of contingency analysis by probabilities - On the relationship between the causal contingency analysis approach and probability theory
13.6. Conclusions - What are the merits of the contingency analysis approach from the perspective of evolutionary economics?
Notes
References
14. The firm as an experimental decision maker
14.1. The rational foundation of the experimentally organized firm
14.2. The nature of entrepreneurial competition
14.3. The nature of business competence capital
14.3.1. Extreme heterogeneity means allocation matters
14.4. Competence bloc theory - The birth, the life, and the death of businesses
14.4.1. Competent customers make a difference
14.4.2. Commercialization and scale up
14.4.3. Receiver competence
14.5. Coase revisited - The firm as an experimental decision maker
14.5.1. The competence bloc integrates the short and the longer term
14.5.2. The firm as a competent team and experimental decision maker
Notes
References
15. Evolutionary economics, routines, and dynamic capabilities
15.1. Introduction
15.2. Can routines do it all?
15.3. How much room for entrepreneurship?
15.4. What great entrepreneurial management looks like
15.5. Entrepreneurial managers in the dynamic capabilities framework
15.6. Dynamic capabilities: Evolution with design, purpose, and strategy
15.6.1. Contributions to dynamic capabilities by evolutionary economics
15.6.2. Narrow view of innovation and change in evolutionary economics
15.6.3. Dynamic capabilities and evolutionary economics reimagined
15.7. Conclusion
Acknowledgments
Notes
References
16. Routines
16.1. The role of routines in evolutionary economics
16.2. Selection
16.2.1. Beyond genotypes and phenotypes: Routines as replicators and firms as interactors
16.2.2. Selection of firms and selection of routines: 'Natural' selection and managerial ('artificial') selection
16.3. Retention (inheritance)
16.3.1. What is being retained?
16.3.2. The retention (inheritance) of routines
16.4. Variation of routines
16.5. Variation, selection, and retention, and their interaction
16.6. Conclusion
Note
Bibliography
17. Organizational routines
17.1. Introduction
17.2. Story and foundations of organizational routines debate
17.2.1. Cognition, problem solving, and routines
17.2.2. Docility a prerequisite for problem solving and organizational memory
17.2.3. Nelson and Winter's observation of organizational routines
17.3. Opening the black box and introducing action into the debate
17.3.1. Replication and emergence of new routines
17.3.2. The role of artefacts as mediators between skills and routines
17.4. The need of entrepreneurial actions and organizational dynamics in face of environmental degradation
17.4.1. New patterns to build and new challenges to handle: The case of environmental degradation
17.4.2. Solving problems differently and building new routines through creative actions
17.5. Conclusion
References
18. Memes
18.1. Introduction
18.2. Cultural evolution, imitation, and the meme's eye view
18.3. Memes as information and instruction
18.4. No meme is an island: Why interconnection is key
18.5. Is everything a remix? Creativity and innovation from a memetic perspective
18.6. Summary and conclusion
Notes
References
19. The path dependence of knowledge and innovation
19.1. Introduction
19.2. Path dependence in the generation of knowledge
19.3. Path dependence of the creative response
19.4. Path dependence and the persistence of innovation
19.5. Conclusions
Notes
References
20. Evolutionary consumer theory
20.1. Introduction
20.2. Modelling bounded rationality
20.3. Accounting for consumption laws
20.4. Insights from the population level
20.5. Demand and innovation
20.6. Consumer demand and structural change
20.7. The coevolution of demand and supply
20.8. Conclusion
Notes
References
21. Evolutionary price theory
21.1. Prices in orderly markets
21.2. Prices in motion
21.3. Prices in the economy as a whole
21.4. Conclusion
Notes
References
22. The coevolution of innovation and demand
22.1. Introduction
22.2. Conceptual background
22.2.1. Co-evolution and economic development
22.2.2. Innovation and economic development
22.2.3. The evolution of demand since the Industrial Revolution
22.3. Coevolution and economic development by the creation of new sectors
22.4. Summary and conclusions
Notes
References
Part II: Evolutionary economic policy and political economy
23. Evolutionary economic policy and competitiveness
23.1. Introduction
23.2. Basic considerations
23.2.1. Qualitative change and growth 'beyond GDP'
23.2.2. Krugman's critique
23.3. Ontology of change: Micro, meso and macro
23.4. System functions: The logic of public intervention
23.4.1. Rationalities of failure
23.4.2. Ability to evolve
23.5. Fitting the pieces: An integrated classification
23.6. Summary and conclusions
Notes
References
24. Smart specialisation
24.1. Introduction
24.2. A short history of S3
24.3. Concepts
24.3.1. Economic geography of innovation
24.3.1.1. Differentiation
24.3.1.2. Concentration
24.3.1.3. S3 and the new geography of innovation
24.3.2. Policy design: Planning and entrepreneurial discovery process
24.3.3. S3 involves several discontinuities relative to usual regional innovation policies
24.4. Fundamentals of S3
24.4.1. Five principles
24.4.2. A simple process
24.4.3. EDP process and outcomes: Transformational road maps and transformative activities
24.4.3.1. Strategic complementarities
24.4.3.2. Spillovers
24.4.3.3. Aggregation level
24.4.4. S3 aims at boosting both vitality and inclusion
24.4.5. The 5 Ds
24.5. Conclusion
Notes
References
25. Evolutionary economic geography and policy
25.1. Introduction
25.2. EEG and smart specialization policy
25.3. Design and implementation of the smart specialization policy
References
26. Global knowledge embeddedness
26.1. Introduction
26.2. Knowledge diffusion through collaboration and mobility
26.3. Drivers of international mobility and collaboration
26.3.1. Determinants of scientist mobility
26.3.2. Determinants of international collaboration
26.4. The global network of knowledge embeddedness
26.4.1. From individual interactions to a global structure
26.4.2. Patterns and dynamics of the global knowledge network
26.4.3. Determinants of country embeddedness
26.4.4. Embeddedness and performance
26.5. Conclusion
References
27. Macro-evolutionary modelling of climate policies
27.1. Introduction
27.2. Macro-evolutionary models
27.3. Macro-evolutionary modelling of climate change and policies
27.4. Research gaps and issues for future modelling
27.5. Climate policy under bounded rationality
27.6. Distributional issues
27.7. Input-output structure and resource scarcity
27.8. Conclusions
References
28. The visible hand of innovation policy
28.1. Introduction
28.2. Three essential elements of AI
28.3. AI as a game changer for innovation policy
28.4. Legitimacy of stakeholders and ethical guidelines for AI
28.4.1. Legitimacy of stakeholders and outsourcing of ethical issues of AI to expert councils
28.4.2. Ethical guidelines of expert councils have numerous problems
28.5. AI is more than a general purpose technology
28.6. Deep learning driving AI endangers human involvement in decision processes
28.7. Using the visible hand of innovation policy at the interface of human and artificial intelligence
28.7.1. The visible hand ensuring the legitimacy of stakeholders
28.7.2. The visible hand changing direction in AI development
28.7.3. The visible hand as guardian of human involvement in the era of deep learning
28.8. Conclusions
References
29. Generalized rules, Nelson-Winter routines, and Ostrom rules
29.1. Introduction
29.2. The generalized rule approach and rules as deductive formats
29.3. Nelson-Winter routines as outward expression of social rules
29.4. Deductive formats in E. Ostrom's analytical architecture
29.5. Discussion
29.6. Conclusion
Note
References
30. Democracy as an evolutionary process
30.1. Introduction
30.2. Democracy and its paradoxes
30.2.1. The representativeness vs governance paradox
30.2.2. The expertise vs direct participation paradox
30.2.3. Novelty vs retained practice paradox
30.3. Democracy as an evolutionary process: Utopia competition
30.3.1. Citizen payoff
30.3.2. Intra-subsystemic evolution
30.3.3. Inter-subsystem dynamics and co-evolution
30.3.4. Emergent properties
30.4. Concluding remarks
Acknowledgments
References
Appendix
Micro-foundations
A formal analysis for intra-subsystem dynamics
31. Public entrepreneurship in economic evolution
31.1. Introduction
31.2. Public entrepreneurship: A brief overview
31.3. The political economy of public entrepreneurship
31.4. Public entrepreneurship and economic evolution
31.5. Summary
References
32. Evolutionary political economy
32.1. Introduction
32.2. Political economy, evolution and transformation
32.3. Models, maps and agent-based simulation
32.4. Concluding remarks
Notes
References
33. Division of labor as co-evolutionary process of ecology, technology, culture, organization, and knowledge
33.1. Introduction
33.2. The true face of Adam Smith and the trade-off between stability and complexity
33.3. Competition and pricing mechanism: Efficiency, risk, uncertainty, and adaptability
33.4. Price, structure, and knowledge
33.5. Changing nature of money in history, from Smith to MMT
33.6. Evolution of organization and classification of firms
33.7. Culture and institutions
33.8. Issues in science philosophy: Falsifiability and ongoing evolution
33.9. From division of labor to co-ordination of innovation
33.10. Conclusion
Acknowledgment
References
34. Evolutionary economics and LDCs: An African perspective
34.1. Introduction
34.2. Economic development in low-income countries: An evolutionary perspective
34.3. Windows of opportunity for transformative change in Kenya and Rwanda
34.3.1. Digital and energy transformation in Kenya and Rwanda
34.4. Policies for transformative change
34.4.1. Governance and regulatory environment for the digital economy
34.4.2. Infrastructural investments for digital communications and renewable energies
34.4.3. Policies for skills development
34.4.4. Multi-stakeholder support for the entrepreneurship ecosystem
34.5. Concluding remarks
BIO Jan Fagerberg
BIO Erika Kraemer-Mbula
BIO Edward Lorenz
Notes
References
35. Globalization and its governance in an evolutionary perspective
References
Index

Citation preview

ROUTLEDGE HANDBOOK OF EVOLUTIONARY ECONOMICS

While dating from post-Classical economists such as Thorstein Veblen and Joseph Schumpeter, the inception of the modern field of evolutionary economics is usually dated to the early 1980s. Broadly speaking, evolutionary economics sees the economy as undergoing continual, evolutionary change. Evolutionary change indicates that these changes were not planned, but rather were the result of innovations and selection processes. These often involved winners and losers, but most importantly, they resulted in actors learning what was and was not working. Evolutionary economics, in contrast to mainstream economics, emphasises the relevance of variables such as technology, institutions, decision rules, routines, or consumer preferences for explaining the complex evolutionary changes in the economy. In so doing, evolutionary economics significantly broadens the scope of economic analysis, and sheds new light on key concepts and issues of the discipline. This handbook draws on a stellar cast list of international contributors, ranging from the founders of the field to the newest voices. The volume explores the current state of the art in the field of evolutionary economics at the levels of the micro (e.g. firms and households), meso (e.g. industries and institutions), and macro (e.g. economic policy, structure, and growth). Overall, the Routledge Handbook of Evolutionary Economics provides an excellent overview of current trends and issues in this rapidly developing field. Kurt Dopfer is Professor Emeritus at the University of St. Gallen, Switzerland. Richard R. Nelson is Professor Emeritus at Columbia University, New York, USA. Jason Potts is Professor at RMIT, Melbourne, Australia. Andreas Pyka is Professor at University Hohenheim, Stuttgart, Germany.

ROUTLEDGE HANDBOOK OF EVOLUTIONARY ECONOMICS

Edited by Kurt Dopfer, Richard R. Nelson, Jason Potts, and Andreas Pyka

Designed cover image: © Getty Images First published 2024 by Routledge 4 Park Square, Milton Park, Abingdon, Oxon OX14 4RN and by Routledge 605 Third Avenue, New York, NY 10158 Routledge is an imprint of the Taylor & Francis Group, an informa business © 2024 selection and editorial matter, Kurt Dopfer, Richard R. Nelson, Jason Potts and Andreas Pyka; individual chapters, the contributors The right of Kurt Dopfer, Richard R. Nelson, Jason Potts and Andreas Pyka to be identified as the authors of the editorial material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library ISBN: 978-0-367-02568-7 (hbk) ISBN: 978-1-032-53339-1 (pbk) ISBN: 978-0-429-39897-1 (ebk) DOI: 10.4324/9780429398971 Typeset in Times New Roman by MPS Limited, Dehradun

CONTENTS

List of contributors

ix

Evolutionary economics: A navigational guide Kurt Dopfer, Richard R. Nelson, Jason Potts, and Andreas Pyka

1

PART I

Foundational issues and theoretical domains 1 Joseph A. Schumpeter: One of the founders of evolutionary economics Heinz D. Kurz 2 Thorstein Bunde Veblen: A founder of evolutionary economics Helge Peukert 3 The foundational evolutionary traverse of Richard R. Nelson and Sidney G. Winter Isabel Almudi and Francisco Fatas-Villafranca

9

11

30

41

4 F.A. Hayek and evolutionary Austrian economics Viktor J. Vanberg

69

5 Kenneth Boulding’s contribution to evolutionary economics Stefan Kesting

79

v

Contents

6 Evolutionary economics and psychology: Where we are, where we could go Brendan Markey-Towler 7 Evolutionary cultural science Carsten Herrmann-Pillath

89

98

8 Evolutionary economics and economic history Andreas Resch 9 Why an evolutionary economic geography? The spatial economy as a complex evolving system Ron L. Martin and Peter J. Sunley

107

117

10 Darwin’s ideas and their mixed reception in evolutionary economics Gabriel Yoguel and Verónica Robert

136

11 Computational evolutionary economics: Minimal principle and minimum intelligence Shu-Heng Chen

147

12 Evolutionary modeling and the rule-based approach Thomas Grebel 13 Contingency in evolutionary economics: Causality and comparative analysis Marco Lehmann-Waffenschmidt Thomas Grebel

161

174

14 The firm as an experimental decision maker Gunnar Eliasson

185

15 Evolutionary economics, routines, and dynamic capabilities David J. Teece

197

16 Routines Markus C. Becker

215

17 Organizational routines Nathalie Lazaric

226

18 Memes Michael P. Schlaile, Walter Veit, and Maarten Boudry

235

vi

Contents

19 The path dependence of knowledge and innovation Cristiano Antonelli and Pier Paolo Patrucco

249

20 Evolutionary consumer theory Andreas Chai and Zakaria Babutsidze

261

21 Evolutionary price theory Harry Bloch

275

22 The coevolution of innovation and demand Pier Paolo Saviotti

284

PART II

Evolutionary economic policy and political economy

297

23 Evolutionary economic policy and competitiveness Michael Peneder

299

24 Smart specialisation Dominique Foray

316

25 Evolutionary economic geography and policy Ron Boschma

332

26 Global knowledge embeddedness Holger Graf and Martin Kalthaus

341

27 Macro-evolutionary modelling of climate policies Karolina Safarzynska

359

28 The visible hand of innovation policy Uwe Cantner and Claudia Werker

369

29 Generalized rules, Nelson-Winter routines, and Ostrom rules Georg D. Blind

381

30 Democracy as an evolutionary process Isabel Almudi and Francisco Fatas-Villafranca

390

31 Public entrepreneurship in economic evolution Jan Schnellenbach

402

vii

Contents

32 Evolutionary political economy Manuel Scholz-Wäckerle

411

33 Division of labor as co-evolutionary process of ecology, technology, culture, organization, and knowledge Ping Chen

421

34 Evolutionary economics and LDCs: An African perspective J. Fagerberg, E. Kraemer-Mbula, and E. Lorenz

433

35 Globalization and its governance in an evolutionary perspective Pascal Petit

445

Index

451

viii

CONTRIBUTORS

Isabel Almudi, Department of Economic Analysis, University of Zaragoza, Spain, RMIT University-Melbourne (Australia), BIFI Institute-University of Zaragoza. Cristiano Antonelli, Dipartimento di Economia e Statistica, Università di Torino and BRICK, Collegio Carlo Alberto, Torino, Italy. Zakaria Babutsidze, SKEMA Business School, Université Côte d’Azur, GREDEG, Valbonne, OFCE, Sciences Po, Paris, France. Markus C. Becker, The Strategic Organization Design Unit, University of Southern Denmark. Ron Boschma, Department of Human Geography and Spatial Planning, Utrecht University, the Netherlands, UiS Business School, University of Stavanger, Norway. Maarten Boudry, Department of Philosophy and Moral Sciences, Ghent University. Uwe Cantner, Friedrich Schiller University Jena, Department of Economics and Business Administration, University of Southern Denmark, Department of Marketing and Management, IT^3, Campusvej 55, Odense, Denmark. Andreas Chai, Griffith Business School, Gold Coast Campus, Griffith University, Brisbane, Australia. Ping Chen, Nat. School of Development, Peking University, and China Institute, Fudan University, Shanghai, China.

ix

Contributors

Shu-Heng Chen, AI-ECON Research Center, Department of Economics, National Chengchi University, Taipei, Taiwan. Jan Fagerberg, Center for Technology, Innovation and Culture (TIK), University of Oslo, Norway. Francisco Fatas-Villafranca, Department of Economic Analysis, University of Zaragoza, Spain. Dominique Foray, Collège du Management, École Polytechnique Fédérale de Lausanne, Lausanne, Switzerland. Thomas Grebel, Economics Department, Technische Universit¨at Ilmenau, Germany. Stefan Kesting, Leeds University Business School, Leeds, UK. Heinz D. Kurz, Graz Schumpeter Centre, Department of Economics, University of Graz, Graz, Austria. Nathalie Lazaric, Université Côte d’Azur, CNRS, GREDEG, School of Business, Economics and Law, France, University of Gothenburg, Sweden. Edward Lorenz, University of Côte d’Azur, France, University of Johannesburg, College of Business and Economics, South Africa. Ron L. Martin, FBA FAcSS FRGS FeRSA, University of Cambridge, UK. Erika Kraemer Mbula, College of Business and Economics, University of Johannesburg, South Africa. Pier Paolo Patrucco, Dipartimento di Economia e Statistica, Università di Torino and BRICK, Collegio Carlo Alberto, Torino, Italy. Pascal Petit, Centre National de la Recherche Scientifique, CEPN, and University Sorbonne Paris Nord, France. Andreas Resch, Institute of Economic and Social History, WU Vienna University of Economics and Business, Vienna, Austria. Karolina Safarzynska, Faculty of Economic Sciences at the University of Warsaw, Poland. Pier Paolo Saviotti, Institute of Economics, Sant’Anna School of Advanced Studies, Italy, GREDEG CNRS, Sophia Antipolis, France.

x

Contributors

Michael P. Schlaile, Research Area 2 “Land Use and Governance”, Leibniz Centre for Agricultural Landscape Research (ZALF), Müncheberg, Germany. Manuel Scholz-Wäckerle, Department of Socioeconomics, Vienna University of Economics and Business, Vienna, Austria. Peter J. Sunley, FAcSS FeRSA, University of Southampton, UK. Walter Veit, Department of Philosophy, University of Bristol, UK. Claudia Werker, Delft University of Technology, Department of Technology Policy Management, Delft, The Netherlands, and RWTH Aachen University, Germany.

xi

EVOLUTIONARY ECONOMICS A navigational guide1 Kurt Dopfer, Richard. R. Nelson, Jason Potts, and Andreas Pyka

The past decades have seen a massive expansion of theoretical, empirical, and methodological publications assembled under the umbrella of “evolutionary economics” (EE). The growth of publications has been paralleled by the introduction of new journals, initiatives for new scientific associations and conferences, and arguably by an increase in the general awareness of the topics dealt with by economists of this camp. Written by an international group of experts in diverse fields, this Handbook provides essential insights into the nature, scope, and various research areas of evolutionary economics. The contents of the Handbook are structured broadly into six topical areas. • The first addresses foundational issues introducing the pioneers of the field and discussing basic issues of theory construction; • the second area revolves around boundary issues and the role of evolutionary sister sciences, such as evolutionary economic psychology, geography, and history, and it lays the common ontological and methodological ground for evolutionary theorizing in economics; • the third deals with behavior, strategies, capabilities, routines of individuals and socially organized units such as firms posited in a co-evolving environment; • the fourth area addresses topics of evolutionary macroeconomics (EME), analyzing the economy as a multi-level knowledge system. Drawing from findings of microeconomics and works on evolutionary trajectories, path dependence, and industry sectors, EME seeks to construct a theoretical edifice of the multi-level economy providing a baseline for evolutionary economic policy; • the fifth area deals with evolutionary economic policy (EEP) addressing policy issues at all levels of the economy as an evolving knowledge system. EEP aims at integrating diverse economic policy areas and at advancing policy proposals apt to cope with timely problems, like those related to the digital revolution and ecological crisis, and, finally; • the sixth topical area discusses pertinent issues of evolutionary political economy (EPE) as a vehicle for coping with future global and long-run challenges. EPE analyses the systemic nexus between the economic and political system, and by expanding the space-time scope it investigates the global economy as the locus of geopolitics conducted in a changing global knowledge and global ecological space. DOI: 10.4324/9780429398971-1

1

Kurt Dopfer et al.

By the beginning of the 20th century, Joseph A. Schumpeter (Heinz D. Kurz) and Thorstein B. Veblen (Helge Peukert) have contended that the very nature of a capitalist economy is its dynamic and that modern economic theory has to take account of this accordingly. Both Veblen and Schumpeter criticized the received doctrine, though their empirical perception and line of theoretical reasoning were different. While Schumpeter proposed to complement extant static theory with a new branch of economic dynamics, thus completing the discipline´s theoretical program, Veblen entertained an anthropological view suggesting borrowing from Darwinian biology and expanding the boundaries of economics into cultural science. Drawing on that tradition, Richard R. Nelson and Sidney G. Winter in their trailblazing “An Evolutionary Theory of Economic Change” (1982) have made a major contribution to what is considered Modern Evolutionary Economics today (Isabel Almudi and Francisco Fatas-Villafranca). They introduced a general evolutionary theory into economics that has a (Neo-) Schumpeterian core with a Veblenian touch. The nexus between evolution and self-organization was explored early on by Carl Menger and other proponents of what became the Austrian school. Following that tradition, Friedrich A. von Hayek made a seminal contribution to solving the problem of spontaneous coordination in a decentralized economy (Viktor Vanberg). In doing so, Hayek demonstrated the twin character of self-organization and evolution. From a broader perspective, economists such as Kenneth Boulding contributed to the construction of the General System Theory which, in turn, served as a platform for constructing a theory of the economy as an open evolving complex system (Stefan Kesting). The economic system view applies to the study of the coordination and emergent structure of the economy on the one hand and its innovation-driven evolutionary dynamic on the other hand. The topical themes along these lines represent the core program of evolutionary economics. Yet, the boundaries of EE are fuzzy and permeable. Significantly, the program may be expanded to connect with sister disciplines aimed at coping with theoretical issues that escape analysis conducted within a more narrowly construed setting. There is a forking of the boundaries in evolutionary economic analysis into “narrow” and “broad,” yet both perspectives are part and parcel of the research program of evolutionary economics. The second topical area includes important contributions that come to be within the broad boundaries of evolutionary economic analysis. The “connecting disciplines” dealt with include Evolutionary Economic Psychology (Brendan Markey-Towler), Evolutionary Cultural Science (Carsten Herrmann-Pillath), Evolutionary Economic Geography (Ron L. Martin and Peter J. Sunley), and Evolutionary Economic History (Andreas Resch). Various ontological approaches including Generalized Darwinism have been proposed seeking to put evolutionary economics on sound paradigmatic foundations (Gabriel Yoguel and Verónica Robert). A host of new methodological challenges arose owing to the premise of an evolutionary approach that both quantity and quality matter and that time asymmetry (rather than conventional Newtonian time) require dealing in equal measure with the historicity of an irrevocable past and an uncertain future. Conceptual and methodological solutions with agent-based evolutionary computational economics (Shu-Heng Chen) and rule-based evolutionary economic modeling combining computational techniques with theoretical reconstruction (Thomas Grebel), and the formalization of contingency and causality for comparative evolutionary economic analysis, signify the new methodological arena (Marco Lehmann-Waffenschmidt). 2

Evolutionary economics

The third topical area deals with the micro-foundations of evolutionary economics. The analytical micro unit is a carrier of cultural knowledge for carrying out economic activities. The elementary components of evolutionary microeconomics are carrier, knowledge, and activities. The ongoing activities of an economy may be viewed either as operations under a given knowledge base of a carrier, such as a firm, or, as referring to the knowledge base itself. In contrast to received economic approaches, evolutionary economics, studies economic operations, like production, consumption and transactions, always in conjunction with the evolutionary dynamic of the knowledge base. Hence, the study of knowledge for economic operations is at the heart of evolutionary economic analysis. A micro-unit as a knowledge carrier may be either an individual or a socially organized unit, such as a firm or household. In evolutionary economics, carriers are understood to be characterized by heterogeneity, and to critically depend on their embeddedness as agents and as agencies. The knowledge dynamic of a micro-unit such as a firm may be captured by a microtrajectory composed of three phases: origination of an information “bit” or a rule variant, its adoption by a carrier, and ultimately retention for recurrent operations. A novel variant such as a business model, decision rule, heuristic, algorithm, organizational design, technological blueprint, or social norm, is a “recipe” (a metaphor used by Boulding, GeorgescuRoegen, Nelson, and others) or analytically, a rule. The knowledge dynamic originates in an invention, that is, the generation of a novel rule. Contemporary economists following an evolutionary approach have traditionally identified production rather than consumption as the principal locus of novelty. The study of the modern firm, informed by findings of the management sciences, set the tone of the research of present-day evolutionary microeconomics. The firm is viewed as a “competent team” organized as an “experimental system” (Gunnar Eliasson), creating and employing dynamic capabilities that embrace routine-based as well as entrepreneurial capabilities (David J. Teece). In parallel, a large literature revolving around the second and third phases (adoption and retention) – epitomized in the notion of routine – has been coming forth. Once generated, the rules of a firm are adopted by way of a process of routinization accommodating the novel rules in the extant knowledge base (second phase), and are retained as the firm’s routines for recurrent operations (third phase). Much attention in the present literature has been given to Nelson-Winter routines that feature behavioral routines (Markus C. Becker) and organizational routines (Nathalie Lazaric). The broad scope and embrace of the concept of rule and routine are signaled by the transdisciplinary concept of Memes as culturally transmitted instructions homolog to genes (Michael P. Schlaile, Walter Veit, and Maarten Boudry). The fourth topical area deals with evolutionary macroeconomics (EME) addressing the economy as an evolving multi-level knowledge system. Theory construction follows in two steps proceeding from the “one” to the “many”. The first step of theory construction features the concept of population. Firms and consumers as carriers of the same kind of rule are members of a population. The adoption of an information “bit” or rule is, qua idea, not confined to a single agent but other agents may adopt that variant as well. A new rule variant may be adopted to the extent it is decodable (not tacit) and property rights may be claimed by potential adopters. The knowledge growth of a population may be modeled as a trajectory that assumes a phase dynamic homolog to that of the micro trajectory. The growth dynamic of the trajectory originates in innovation and unfolds as a complex process of interactions among 3

Kurt Dopfer et al.

multiple heterogeneous agents. Specifically, the first entry into a market (first phase) by an innovator (not necessarily the inventor), is followed by an interactive process of adoption in a population of adaptive agents in a selective environment (second phase), that settles in a meta-stable process of rule retention with a (typically logistic) adoption frequency curve at maximum (third phase). Producers and consumers entering a market meet there as agents of supply and demand. A market develops as a co-evolutionary process composed of a producer trajectory and a consumer trajectory (effective in that market). At any time the co-evolutionary phase dynamic governs in a particular way the operational dynamic of ongoing market transactions. In neoclassical economics all variables of the co-evolutionary circuit are exogenous and the study of the market is confined to equilibrium analysis at the operational level leaving in limbo the deep level of its structure and evolutionary dynamic. Evolutionary market theory studies the rules governing the behavior of the market agents and the rules (or characteristics) of the products traded. The theoretical scheme includes - as rule carriers - naturally both the producers and the consumers. The analysis of the producers as supply agents investigates issues like decision-making under uncertainty, bifurcations, path dependence, local learning, and diverse adaptive capability in a selective environment (Christiano Antonelli and Pier Paolo Patrucco). The study of the co-evolving consumer trajectory deals with a broad spectrum of topics, such as needs- and wants-discovery, consumer learning, formation of consumer preferences, and the role of consumers as proactive drivers of innovation and change in a market (Andreas Chai and Zakaria Babutsidze). The study of coordination in a partial market then focuses on a co-evolutionary circuit composed of populations of producers and consumers as they unfold their trajectories. The second step from “one” to “many” in macro theory construction is from one population to many populations. The analysis deals with all information “bits” or rules like behavioral routines, cognitive rules, technologies, or social norms on the one hand and with all trajectories of their physical actualization in time and space on the other hand. The subject matter of EME is the study of (i) the coordination and structure of the economy as a system and (ii) the evolutionary process as its continual endogenous change. A “general theory” of evolutionary economics considers all producer and consumer trajectories as determinants of total supply and demand in a market economy. The orthodox canon is preoccupied with aggregates, not with the structure and evolution of macro demand and macro supply. Its theoretical baseline is a product and factor space composed of quantity-price vectors which lack any qualitative attribution required for statements about structure and qualitative change. In contrast, an evolutionary market theory is anchored in a price theory that deals with the price dynamic arising from processes of economic change. Evolutionary prices guide the coordination of knowledge and act as incentives for innovations of producers and consumers at all levels (Harry Bloch). At the micro level, prices influence the behavior of innovators and followers, at the meso level, they are effective in the process of creative destruction and path dependence in a single market, and at the macro level, they function as parameters of spontaneous order (rather than equilibrium) in an economy. In a related vein, long-term economic growth is being modeled as a process of co-evolution between innovative production and adaptive change in demand, stated as dynamic intra- and intersectoral relationships of an economy (Paolo Pier Saviotti). The 5th topical area addresses the domain of evolutionary economic policy (EEP). Economic policy – referring to a nation-state – studies the measures by a body of politics 4

Evolutionary economics

designed to steer the economy into a desired normative direction. Economic policy based on mainstream theory revolves essentially around corrections of economic outcomes. It corrects market failures, redistributes incomes, interferes anti-cyclically, and stabilizes the monetary and financial spheres. It is ex-post, reactive, employing essentially a rather short-term perspective. In contrast, EEP deals with short-term policy issues in equal measure, but it is not confined to them. It uses a framework that posits alongside short-term also mediumterm and long-term variables on the historical time scale. Economic policy in an evolutionary key embraces all levels of the economy as a knowledge system. Multi-level EEP includes enterprise and employment policies (micro), sectoral innovation policies, cluster, network, and regional policies (meso), innovation infrastructure policies, transformation policies, or mission-led policies (macro), all combined in the concept of competitiveness policy that ensures the viability and evolvability of the economy as a whole (Michael Peneder). Particular kinds of generic policies have been suggested by employing novel concepts like “smart specialization strategy” aimed at creating and developing networks of innovators and innovation ecologies (Dominique Foray). Its notions, like entrepreneurial discovery, local capabilities, complementarity, relatedness, and region-specific growth trajectories, link naturally to the sister science of evolutionary economic geography, making it a source for policy analysis (Ron Boschma). As knowledge systems, national economies are part of a global network of knowledge, with scientists and research organizations collaborating at all levels across geographic scales making global science policy a strategic factor of evolutionary economic policy (Holger Graf and Martin Kalthaus). EEP is facing epochal challenges considering the contemporary ecological transformation and digital transformation. Expanding the historical time scale, new explanatory variables have been introduced into the theoretical body providing a basis for transformation policies. Economic climate models have been devised focusing on the co-evolutionary dynamic of producer and consumer trajectories, differential social responses, and distributional results in light of different policies aimed at reducing carbon dioxide emissions (Karolina Safarzynska). While digital transformation promises novel solutions for the most pressing problems of our time, it is also a source of radically new and little-understood problems awaiting policy solutions. As a case in point, artificial intelligence (AI) as a constituency of innovation policy connects with principal ethical challenges calling for a re-assessment of “legitimized stakeholders” and for “ethical guidelines” in the face of shifting decision-making authority between humans and machines due to deep learning (Uwe Cantner and Claudia Werker). The sixth topical area discusses pertinent issues of evolutionary political economy (EPE). This strand addresses traditionally the systemic nexus between the economic and political system on the one hand and the global economy as the domain of geopolitics on the other hand. Economic policy is anchored in the institutions of a political system. Understanding its nature and functioning is paramount to understanding the possibility space of economic policy. Like the economic system, it may be conceived as a rule system allowing us to bridge economic analysis with adjacent strands of political decision-making, public entrepreneurs, and political system analysis. Looking for a systemic nexus, a generalized rule approach reveals close commonalities between the analytical core of Nelson-Winter routines and Elinor Ostrom public decision rules paving the way for a fruitful dialog between scholars of the two complementary fields (Georg D. Blind). Countries sharing the ideals of the European enlightenment have adopted the rule system of democracy as opposed to any system of autocracy. Modern democracy, molded by individual and group digital connectivity, is a highly dynamic 5

Kurt Dopfer et al.

system inviting a Schumpeterian perspective for exploring the nature of its complex dilemmas and paradoxes. (Isabel Almudi and Francisco Fatas-Villafranca). At the micro level, a change agent in the form of a public entrepreneur enters the political arena complementing the familiar rent-seeker of public choice theory (Jan Schnellenbach). The research agenda of EPE can be traced to the political economy of the classical school of economic thought, to the “Original Political Economy” (OPE). In the view of its founders, an economy rests on two pillars: knowledge and nature. Adam Smith considered the division of labor and specialization to be the driver of economic development and a major source of the “Wealth of Nations”. Thomas Malthus and David Ricardo, in turn, highlighted the ecological limitations proposing that marginal returns of food from available arable land constitute the critical determinant of the survival and welfare of a population. The founders of OPE saw the economy as being built on a “knowledge system” and an “ecological system”. The political system of OPE was aligned with the nation-state as it emerged after the breakdown of the Ancien Régime in the 18th and 19th centuries. The economic development of European nation-states propelled by the industrial revolution led to a rapid expansion of trade relations with non-European countries resulting eventually in the oppressive colonial world system. Karl Marx proposed that all human history takes place as a succession of class struggles, and he conceived the world economy as a geopolitical space where an international class struggle determines the fate of nations. In his theory, the economic development of the colonies depends on the geopolitical dynamic of the capitalist world system rather than native factors of knowledge or ecology. The canon of EPE revolves around a reconstruction of OPE under present-day historical conditions. It analyses the world economic system as a geopolitical space that posits the economies of all nation-states or unions of states. The canon shares with Marxian political economy the normative impetus for change, but it does not seek solutions to the pressing problems of our time by waging a class struggle. Instead, it puts forward cooperation to improve the conditions of knowledge growth and ecological sustainability at a world scale. The rapid climate change and loss of biodiversity paralleled by a digital revolution demand entirely new forms of distributive and allocative schemes aligning responsibilities, rights, duties, commitments, and claims with the realities of the new epoch. The epochal transformation process calls for a restatement of national and global schemes of policy desiderata, backed by novel methodological approaches like agent-based modeling, experimentation, simulation, and multi-level analysis (Manuel Scholz-Wäckerle). Smith´s scaled-up division of labor unfolds in our times as a cultural transformation calling for models that allow for non-linearity, non-equilibrium, bifurcations, and systemic tipping points (Ping Chen). An important validation test of EEP and EPE consists in demonstrating their usefulness for theoretically framing the “big” policy problems and for providing a platform for their practical solution. A pressing problem of our times is the low standard of living in poor countries and their low participation in the world economic system. EPE provides a strategic framework that interprets problems of mass poverty and inequality in the context of a deteriorating natural environment, and it proposes solutions by exploiting the full potential of digitization and innovative entrepreneurship (Jan Fagerberg, E. Kraemer, E. Lorenz). On a global scale, EPE suggests that the competition for geo-political zones of influence must give way to global commons governance and a new consciousness conducive to solving planetary problems (Pascal Petit). 6

Evolutionary economics

The topical areas discussed may be seen to demonstrate the enormous research dynamic in EE. However, there is also widespread recognition that there are many lacunae and shortcomings in the cumulated research conducted thus far. For instance, the coevolutionary nexus between the strata of behavior and the product space is underresearched, the nested character of the multi-level system composed of agents, populations, and the economy as a whole is little explored, work on systemic complementarities is wanting, the significance of semantic information for theorizing structure and qualitative change and for devising cladistics taxonomies is not generally acknowledged, mathematical or formal-analytical methods remain to be exploited, the large potential of theory as a scaffold for devising effective evolutionary economic policies is yet to be tapped, and the list, with little effort, may be extended. Despite the current shortcomings, looking at the scientific potential from within the camp, evolutionary economics rests on sound ontological-paradigmatic foundations, employs analytical boundaries wide and permeable enough to cope with all extant and new theoretical problems, and provides, at this stage, the core of a coherent cumulated body of theory that promises exciting developments in the future. It may serve as a reliable basis for coping with the most pressing policy challenges of our time.

Note 1 The “Navigational Guide” aims at both giving an overview of pertinent themes of evolutionary economics and at providing the reader with some hints about where to find a respective source in the Handbook. The names within the brackets refer to the authors who in their entries deal with one or several topics addressed in the text to follow.

7

PART I

Foundational issues and theoretical domains

1 JOSEPH A. SCHUMPETER One of the founders of evolutionary economics1 Heinz D. Kurz

1.1

Introduction

It is fair to say, I think, that amongst economists Schumpeter is today considered, if not the most important, then by far the most prominent representative of an evolutionary approach to economics. This entry seeks to answer the question of what can be meant by calling Schumpeter a founder and representative of evolutionary economics. Several distinguished Schumpeter scholars have dealt with this issue in historiography; it suffices to mention Richard Nelson and Sidney Winter (1982a, 1982b, 2002), Nelson (2018), Geoffrey Hodgson (1993), Stanley Metcalfe (1998, 2005), Kurt Dopfer (2001, 2005, 2016), Dopfer and Jason Potts (2008) and Dopfer and Nelson (2018). These and other contributions are invaluable sources of information and assessment, and people interested in the theme are asked to consult them.2 I approach the problem under consideration in terms of the following steps. In Section 1.2, I sketch in unforgivable brevity some important events in the history of evolutionary theory in biology, with an emphasis on Charles Darwin’s work. Section 1.3 then informs about how Schumpeter received the latter. Whilst full of praise for Darwin’s achievements, Schumpeter felt that an entirely new concept of evolution had to be elaborated that is congenial to the subject matter of economics and other social sciences. Section 1.4 turns to the reception of Darwinian ideas by Friedrich Engels and Karl Marx and Schumpeter’s assessment of it. Section 1.5 compares Marx and Schumpeter’s alternative perspectives on the law of motion of modern society. While Schumpeter denied that history has a predetermined goal, Marx was convinced that it would bring about a classless society. Without too much of an exaggeration, Marx may be said to have been a “Spencerian” evolutionist, whilst Schumpeter was a “Darwinian” one. Schumpeter rejected Marx’s explanation of the necessary demise of capitalism and insisted that as a “method” of inducing and absorbing innovations and economic change it cannot possibly run out of energy unless political measures “fetter” its dynamic forces. However, capitalism might have to give way to socialism not so much because of hard economic “facts”, but because powerful “ideas” generate a political climate that is inimical to it. These ideas, Schumpeter insists, are largely based on a misunderstanding of the modus operandi of capitalism and cause wrong policy and political inferences. By showing that the ideas are mistaken, Schumpeter hopes to contribute his share to turning the tide and cultivates

DOI: 10.4324/9780429398971-3

11

Heinz D. Kurz

a clandestine hope that socialism can still be prevented from spreading. Section 1.6 localises the Schumpeterian “entrepreneur” historically in a period in which labour productivity began to rise significantly due to technological progress. This period was rendered possible only because “cultural entrepreneurs” had paved the way to the Industrial Revolution in Europe by establishing a “culture of growth” (Mokyr). Section 1.7 turns to the role of social conflicts in economic development and the self-transformation of society. It compares the properties of capitalism and socialism as to economic efficiency and to what extent capitalism’s cyclical movement and crisis-proneness are (mis)interpreted as fundamental weaknesses. In this context, Schumpeter’s reservation towards effective demand management and incomes policy is discussed. Section 1.8 turns briefly to how capitalism and socialism fare with regard to culture and democracy. While Schumpeter clearly opts in favour of capitalism, his judgement is not enthusiastic. Section 1.9 contains concluding observations. Before I proceed, it is useful to specify what is meant by “evolutionary economics”. The following definition Kurt Dopfer (2016: 184) proposed serves this purpose very well: The subject matter of evolutionary economics is the economy as an evolving system. The central scientific approach lies in investigating the theoretical nature of the system and that of evolutionary dynamics. The definition covers inter alia the concern of earlier authors with unravelling the “law of motion” of society. This requires understanding the socio-economic system under consideration, the relationships between its parts, and its metabolism with the environment, on the one hand, and an analysis of the factors affecting its development over time from within, endogenously, on the other. This results in a theory of the process of self-transformation of the socio-economic system that is reflected both in quantitative and qualitative changes and in changes in the modus operandi of the system as a whole. The main driving forces of the system are heterogeneous agents, possessed of different motivations, capabilities, and resources, and able to learn from the success or failure of their actions. The main characters that populate Schumpeter’s theoretical world are “entrepreneurs” or “agents of change”, who engage in innovations and trigger processes of “creative destruction”; bankers who provide the funds to carry out innovations; producers and hedonic consumers, who adjust to the novelties and fuel waves of imitation; workers who respond to the changes they are confronted with; intellectuals, who interpret the socio-economic world and form judgements about alternative social orders; and politicians, the public administration and the state who intervene in the economic process. The interests of the acting agents are often antagonistic, which gives rise to conflicts that discharge themselves in terms of a wide spectrum of socioeconomical change, reaching from reforms to political upheavals and revolutions. Schumpeter may be said to have laid the foundation of his evolutionist perspective in his Theorie der wirtschaftlichen Entwicklung (1912). Later works including further editions of his magnum opus, his Business Cycles (1939) and Capitalism, Socialism and Democracy ([1942] 2008) offer essentially two things: first, they elaborate on the foundation and refine some of its concepts; secondly, they apply the theory in interpreting major historical events in terms of a histoire raisonnée of modernity. This concerns, on the one hand, a more narrowly conceptualised economical perspective on long waves of development essentially since the Industrial Revolution and, on the other hand, a much wider perspective on basically all spheres of social life in an analysis of the trend of modern society confronted with the alternative of capitalism and socialism.3 12

Joseph A. Schumpeter

I turn now to a few important signposts in the history of the theory of biological evolution and especially Darwin’s contribution (Mayr 1982, Mokyr 1991, Dopfer 2001, Lorenz 2020). Later it will be shown how these might have inspired evolutionism in economics.

1.2

Charles Darwin and biological evolution

When Charles Darwin (Darwin 1859) published The Origin of Species in 1859, the term “evolution” had already been in use for a while, especially amongst students of palaeontology. However, it meant different and occasionally even contradictory things to different scholars. The following questions soon assumed centre stage in the discussion: can evolution bring about true novelty above and beyond what God had created, provided we are willing to follow the biblical texts on creation (Genesis)? Or does it just recompose a given, never changing finite set of elements? Answering this question in the positive implied a deep conflict with the Catholic Church. The extinction of species, documented abundantly by fossil finds, challenged the religious-cum-essentialist position. A way out of the lingering dilemma was to interpret the obvious disappearance of species as simply a complicated conversion in form that hides the preservation of substance from the eye of the observer. It was especially Darwin, who, based on comprehensive studies of plants and animals during his discovery trip to the Amazonas, Patagonia, and the Galapagos Islands, paved the way to an understanding of evolution as a dynamical process of the development of species. This process revolves around random variations of properties, which are more or less well attuned to the given specific environment. Those properties that are better attuned to it involve advantages for their holders and their selection in a battle of survival. These properties will then be inherited by their offspring. The important point to note is that in this perspective the individual does not react to the environment and its variation, as in Jean-Baptiste de Lamarck’s view, published in his Histoire Naturelle between 1815 and 1822, who had argued that the organs of animals change according to whether they are used or not and that acquired skills are then bequeathed. In Darwin’s different view, the environment selects those individuals possessed of the most advantageous properties. Accordingly, random variation comes first, and “natural selection” by the environment only afterwards. Here, two further events in the fascinating history of the development of evolutionary theory deserve to be mentioned. First, there are the inheritance laws of Johann Gregor Mendel, discovered in the mid-1860s. These assign the inheritance of phenotypical individual properties, “alleles”, to the dominance of certain properties in the genetic material of the carriers (their “genes”). Mendel’s laws allowed one to explain observed variations in the genotype of descendants. Secondly, there is Hubert de Vries’ discovery around 1904 of the phenomenon of “mutation”. Mutations are accidental variations in the genotype itself. Their existence means that alleles are themselves not immutable. Their discovery allowed for the first time in biological evolution to talk about the generation of “novelty” (Lorenz 2020: 13). However, since surviving mutations are relatively rare in nature, the recombination of the genetic material in sexual reproduction continues to be of great importance. The question to which we now turn is what can the social sciences and especially economics learn from evolutionary biology, which ideas and concepts can be carried over to them, and how much does what is being adopted have to be adapted in order to fit into the new environment and improve our understanding of the subject matter? We begin with some reflections on Schumpeter’s reception of Darwin’s doctrine. 13

Heinz D. Kurz

1.3

Schumpeter on Darwin and Darwinian ideas in economics 1.3.1

Schumpeter’s “Monroe doctrine”

There is reason to presume that Schumpeter was familiar with the basics of evolutionary biology from the early time of his academic career onwards, as several references to Darwin and evolutionary biology in his oeuvre document. In the History of Economic Analysis, published posthumously in 1954, he comments in some detail on “Darwinian Evolutionism”. He is full of admiration for Darwin’s Origin of Species and especially of the “Historical Sketch” contained therein, which traces the gradual emergence of its path-breaking ideas. According to Schumpeter, this is “one of the most important pieces of scientific history ever written”. It is of particular interest to him, because it presents a case study about one of the objects of our interests – the ways of the human mind and the mechanisms of scientific advance. In addition, it elucidates a concept that plays some role in our own story [i.e., the history of economic analysis], the concept of Inadequate Acknowledgment of Priorities. (Schumpeter 1954: 444) His praise of Darwin is boundless: In everything he did, this man was a living and walking compliment to himself and also to the economic and cultural system that produced him – a point recommended to the reader whenever he feels like ruminating on the civilization of capitalism (and, incidentally, about more modern forms of organization of research). (ibid: 444 n. 18) However, praising Darwin is one thing, thinking that his insights could somewhat be carried over to economics, is an entirely different thing. Interestingly, while Schumpeter was deeply impressed by Darwin, he was sceptical that economists could learn much from him. In his first major work, Das Wesen und der Hauptinhalt der theoretischen Nationalökonomie, published in 1908, he had already put forward a “sort of Monroe doctrine” (1908: 536), according to which economics stands on its own and is strictly separate from other scientific disciplines, including philosophy, psychology, and also biology. He was particularly critical of the request to provide a psychological foundation for the concept of utility. Other disciplines in the social sciences, he insisted, “can give us only little – or nothing. In the interest of clarity it is imperative to stress their nullity and throw this ballast over board” (1908: 553). These are bold pronouncements by a 25-year-old man, who sees economics a great deal closer to the natural sciences than the humanities: “According to its methodological and epistemological nature, pure economics would be a ‘natural science’ and its theorems ‘natural laws’” (1908: 536). If evolutionary ideas and concepts were to play any role in economics, then the biotope from which they would have to emerge had to be economics itself, and the material out of which they would have to be moulded would have to come from there. This was Schumpeter’s conviction from an early time onwards and, cum grano salis, remained so for the rest of his life. His evolutionism therefore had to be something genuinely novel, elaborated ab ovo, and could at best involve a faint echo of biological evolutionism, originating with Darwin. As we shall see, in it the unintended consequences of human creativity, innovativeness, and, more generally, action play an important role. It was especially the Scottish Enlightenment that had attributed great importance to such consequences. 14

Joseph A. Schumpeter

1.3.2

Evolution – A directionless process

There are further aspects of Schumpeter’s respective view that deserve to be mentioned. In Theorie der wirtschaftlichen Entwicklung (Theory of Economic Development), published in 1912, he rejects all “searching for an objective sense of history and also the postulate … of some development in the sense of a line of development that is to be understood uniformly”. This, he is convinced, is based on a “metaphysical prejudice”. His criticism is not only directed at Georg Wilhelm Friedrich Hegel: “To this belongs also a variant of the idea of development, which is centred in Darwin – at least if this way of looking at things is applied to our field in a simple analogous way” (1912: 89–90). It is obviously Herbert Spencer’s interpretation of evolution and especially social Darwinism that Schumpeter finds difficult to sustain. Spencer had translated Darwin’s concept of “natural selection” as the “survival of the fittest”, a term Darwin used only in passing. Michael Ruse, in a paper on how The Origin of Species was received and his ideas developed, distinguishes between two trends, one “bad” and the other “good”. The bad one he traces back to Spencer who tried to make evolution into a doctrine of progress – from the weak to the strong and from the not so good to the better. The main representative of the good trend is said to have been Darwin himself, to whom evolution was “a directionless process, going nowhere rather slowly” (Ruse 1988: 97). Schumpeter can be said to have shared such a view of the social sciences. The process of socio-economic development is “evolutionary” (Schumpeter 1912: 184): it has neither a predetermined goal (or “equilibrium”) towards which it gravitates nor a predetermined path on which it converges (Schumpeter [1932] 2005). It is a restless process, incessantly kept in travail by the rivalry of agents and the innovations they carry out. We now take a brief look at what Schumpeter thought about the way Friedrich Engels and Karl Marx received Darwin’s findings.

1.4

Schumpeter on Engels and Marx on Darwin

1.4.1 Darwinism – The natural science foundation of class struggle? Under the influence of Friedrich Engels, Marx for a short while seems to have tinkered with the idea of absorbing Darwinian evolutionary biology into political economy. Engels had drawn Marx’s attention to Darwin’s Origin. The book, he had opined, contains the “natural history foundation of our point of view”, and in a letter to Ferdinand Lassalle called it the “natural science foundation of class struggle”. It is also said to have done away once and for all with “teleology” in the natural sciences. Engels interpreted Darwin’s doctrine of the struggle for existence as “simply a transfer of Hobbes’ doctrine of bellum omnium contra omnes and the bourgeois economics of competition and Malthus’ theory of population from society into animated nature”. (See the references provided in Lorenz 2020: 16.) Marx studied Darwin’s Origin and was deeply impressed by it. He saw a parallel between it and Capital in that both books sought to reveal the history of various species and effectively uncovered major principles shaping it in each case. However, Marx soon understood that the “purely accidental” basis of Darwin’s doctrine did not provide any support for his conviction that social development would bring about a classless, exploitation-free society. In commenting on Engels and Marx’s reception of Darwin, Schumpeter confirms the view that these authors also dealt with processes of qualitative change. However, he disputes strongly that Marx’s “class struggle” and Darwin’s “struggle 15

Heinz D. Kurz

for existence” and the principle of “natural selection” are related phenomena: “Marx may have experienced satisfaction at the emergence of Darwinist evolutionism”, Schumpeter opines. “But his own had nothing whatever to do with it” (1954: 445). Marx’s view that history had a predetermined goal flew in the face of Darwin’s conviction that change resulted from random variations, that is, unpredictable events that are highly uncertain in outcome.

1.4.2

A discovery as important as that of the heliocentric system

Schumpeter stresses that he does not “of course presume to judge [Darwin’s] book as a professional performance in its own field” (ibid). His attention focuses exclusively “on the social significance of the book and on its significance for the social sciences” (ibid: 444). In this regard, he opines, the harvest is meagre as also Marx understood after some deliberation. This does not affect in the least Schumpeter’s judgement that the Origin and then Darwin’s Descent of Man (1871) provided “one of the biggest patches of color in our present picture of that period’s Zeitgeist”. Darwin’s achievement is said to be of “secular importance for mankind’s cosmic conceptions”, comparable with the discovery of the “heliocentric system” (ibid: 445). Marx had understood that evolutionary biology did not offer ready-made concepts and analytical tools that had only to be carried over to political economy and put to productive use there. What it offered, however, was a standpoint that provided a most fascinating perspective of economy and society as an evolving, self-transforming system. This was in stark contrast to, and implied a radical break with, doctrines, old and new, that conceived the socio-economic system in analogy to mechanical devices or machines, such as clocks, or the universe, that is, entities that were taken to run according to preconceived programmes. In economics, the static, mechanical point of view was championed by Alfred Marshall (1890), who had famously chosen for his Principles of Economics the motto: “natura non facit saltum”, implying continuity and ruling out disruptive changes.4 But the socioeconomic system was all the time in travail from within. The mechanical analogy was utterly misleading. Its essentially static point of view knows no learning of agents, no need to adapt preferences to new goods or better qualities of known goods and the disappearance of old goods, no need to learn how to operate new machinery, and so on. With regard to dynamic, evolutionary systems, it is precisely the processes of learning, selection, adoption, adaptation, non-incremental change, etc. that matter. Marx deserves credit for having made an early attempt to tackle such issues. With his claim that capitalism, like all preceding modes of production, is a transitory form of organising the production and distribution of wealth that will inevitably be replaced by socialism, Marx had confronted evolutionary thinking with a major challenge. Several Austrian economists, including Böhm-Bawerk, Ludwig von Mises, Friedrich August Hayek, and Schumpeter, accepted the challenge and were keen to refute the claim. “Capitalism vs. Socialism” dominated much of the discussions and often heated debates towards the end of the 19th century and during the first half of the 20th century, both in the social sciences and the public sphere. Schumpeter also forcefully weighed in on the controversy, but his point of view differs markedly from that of others.5 He neither shared Marx’s position nor that of any of the Austrian economists. Socialism was clearly not only possible. But was its extension worldwide inevitable, as “scientific socialism” contended? 16

Joseph A. Schumpeter

1.5

Making his sense of Marx

1.5.1 Marx interpreted in a “conservative sense” Marx’s oeuvre is by far the most important source of inspiration of Schumpeter as regards his understanding and analytical conceptualisation of the process of creative destruction and the self-transformation of society (see Kurz 2012, 2018a). This process typically takes off from the economic sphere, but then radiates to all other spheres of social life and back again to its origin. Schumpeter does not simply adopt what he finds in Marx. His way of dealing with “Old Moor” is clearly spelled out in the following statement: “To say that Marx, stripped of phrases, admits of interpretation in a conservative sense is only saying that he can be taken seriously” ([1942] 2008: 58). In this sense, he takes Marx very seriously and is constantly in search of moot points in his analysis that allow him to put upside down the thrust of his reasoning. This exercise is designed to show that capitalism is not inferior to socialism and that arguments of its critics, especially the “intellectuals”, are essentially based on misconceptions. A central passage in Capitalism, Socialism and Democracy reads: “The masses have not always felt themselves to be frustrated and exploited. But the intellectuals that formulated their views for them have always told them that they were, without necessarily meaning by it something precise” ([1942] 2008: 26). It was Marx who had rendered a precise meaning to the phrase, giving it a quasi-scientific veneer. Defending capitalism therefore presupposes in the first place showing that Marx’s doctrine cannot be sustained.

1.5.2 “Law” of the falling rate of profit According to Marx, the fact that capitalism is unsustainable in the long run follows from the “law” of the falling tendency of the general rate of profit. The linchpin of this “law” is Marx’s conviction that a particular form of technological progress is congenial to the capitalist mode of production. With an increasing social antagonism between labour and capital, capitalists seek to contain the power of the working class by introducing new techniques that replace “living” labour by “dead” labour, incorporated in capital goods (tools, machinery, etc.). Since living labour is the source of surplus value and profits, the source is taken to gradually dry up. A falling general rate of profit expresses the fact that capitalism loses its vitality and will eventually get replaced by socialism. “Scientific socialism” (Engels) is to be credited with demonstrating this with an authority that is typically attributed only to the natural sciences. The law is seen to result from profit-seeking behaviour, to which competitive conditions compels capitalists. In order to survive in the market, they are continuously on the lookout for cost-reducing methods of production that allow them to prevail over their “inimical brothers”. The irony of Marx’s argument is that individually rational behaviour turns out to be collectively irrational, since it undermines the capitalist social order and eliminates the class of capitalists.

1.5.3

The “capitalist engine”

Schumpeter is not convinced by Marx’s explanation.6 First, profits cannot generally be traced back to the “exploitation” of workers: they rather typically result from the productivity enhancing effects of innovations that increase the product and need not diminish workers’ share in it. Secondly, capitalism suffers neither from a decline in the long-term rate of technological progress, which would be accompanied by a falling profitability and a 17

Heinz D. Kurz

slowing down of economic development and growth, nor from a persistent bias of technological progress. As regards the first part of the argument, Marx apparently mistook the downward phase of a long wave of economic development – a Kondratieff cycle – for the whole. Had he focused attention on the upward phase, he would have had to enunciate a law of a rising tendency of the general rate of profit. As a system, whose characteristic feature is to generate and absorb continually novelty and change, Schumpeter surmises, capitalism simply cannot be thought of as running out of steam: “we may put some trust in the ability of the capitalist engine to find or create ever new opportunities since it is geared to this very purpose” (Schumpeter [1942] 2008: 117). As regards the claim that technological progress in capitalism is persistently biased in a particular way, Schumpeter insists that economic development is path dependent, which implies that past waves of technological change impact to some extent on later waves. For example, if technological progress was of a labour saving-cum-capital using form for some time, this would sooner or later be reflected in falling or only slowly rising real wages, which would in turn provide an incentive to develop inventions that use relatively more labour and economise on factors that have become more expensive. Therefore, there is no presumption that forms of technological progress reflected in a rise in the “organic composition of capital” can be expected to persist throughout capitalist history. According to Schumpeter, the “facts” to which Marx alluded cannot bear the brunt of his claim that capitalism is doomed to failure. Schumpeter nevertheless does not rule out this possibility, not necessity, but his explanation is entirely different and revolves around players that are either absent or at best only vaguely discernible in other authors. While Marx focuses attention on the conflict between workers and capitalists and marginalist authors contemplate the utility maximising homo oeconomicus, Schumpeter stresses the importance of the carriers of new ideas and the real consequences these have. On the one hand, there are entrepreneurs – innovators or “agents of change” – who brush aside received constraints and widen the socio-economic space in terms of new products, new methods of production, new forms of organization of firms, new institutions, and so on. On the other hand, there are intellectuals – political activists, philosophers, and scientists – who interpret the complex world and assess alternative social orders. The former lead, seduce, or even force the “hedonic” majority to want new goods and teach them how to use them, or they provide strong incentives to producers to introduce new methods of production that allow one to reduce costs and increase profits. The latter rally people behind political movements that seek to defend or overcome the received social order. The generation of new ideas by the homo inventivus and their realisation by the homo innovativus trigger population dynamics that have multiple effects and transform the economy, society, and culture. The homo intellectualis then passes judgements on whether these are desirable or not.

1.6

Ideas have consequences 1.6.1

Ideas vs. facts

Since Marx’s argument is not compelling, how does Schumpeter explain the possible fall of capitalism? Might “ideas” bring it about? Numerous writers have since a long time insisted that ideas impact the evolution of the socio-economic system (Weaver 1948, Gehrke 2019). By shaping people’s motives and actions, they shape the world in which we live. Schumpeter may be said to have put forward a particular version of this view that revolves around the

18

Joseph A. Schumpeter

two different types of agents mentioned: entrepreneurs on the one hand and intellectuals and political activists on the other. While much of his attention focuses on the economic sphere, he insists that entrepreneurship is not only encountered there, but also in the arts and sciences and, though perhaps to a lesser extent, in politics and public administration. While innovations often start in the economic sphere, they then radiate to other spheres of society, where they call forth adjustments and induce further innovations, which often retroact to the economic sphere. These processes are cumulative and self-enforcing and give rise to particular dynamics, which reflect the specificities of cause, recipients, and propagators involved. The world in which we live is characterised by a high and growing complexity. A deepening social division of labour has brought forth inter alia people that specialise in interpreting what is poorly understood and provide guidance – Adam Smith spoke of “philosophers and men of speculation” (WN I.i.9). These study the properties and dynamic features of economy and society and whether they can be influenced for the better. They produce social theories, political doctrines, and Weltanschauungen and shape the ideological outlook of people.7 Given the complexity mentioned, intellectual bubbles, contagion, and herd behaviour are omnipresent. They are supported by the fact that capturing other people’s minds pays. Not only politicians know this perfectly well, but also the bosses of firms, whose customers may get addicted to the products and services they sell. When did the Schumpeterian entrepreneur, conceived as innovator, assume a leading role in shaping modern economic history? The following interpretation provides a provisional answer to this question with reference to (i) Marx’s distinction between the production of “absolute” and of “relative surplus value” (Marx [1867] 1954: 299), (ii) Max Weber’s work on the Protestant ethic and the “spirit” of capitalism (Weber [1930] 2001), and (iii) Joel Mokyr’s reconstruction of the evolution of a “culture” of innovation and growth as the non-intended consequence of numerous institutional innovations in the 16th and 17th centuries in Europe (Mokyr 2017). Marx’s distinction refers broadly to the time before and after the turning point marked by the beginnings of the (First) Industrial Revolution; Weber focuses on the early period in which technological innovations were modest and economic growth was driven by high savings and investment; Mokyr deals with the achievements of “cultural entrepreneurs” who prepared the ground for the economic take-off towards the end of the 18th century; Schumpeter’s economic entrepreneur finally benefited from a situation, in which the spirit of capitalism was still present, although in a somewhat subdued form, and a culture of innovation cushioned his efforts. A few remarks must suffice.8

1.6.2

The “spirit” of capitalism

In an essays published in German in 1905–1906, which Schumpeter studied with great attention, Max Weber had investigated what happened subsequent to the Reformation, with the “Calvinist and other Protestant sects paving the way to the rise of the ‘spirit’ of capitalism”, that is, the “utopia of a ‘capitalist’ culture, i.e., one dominated solely by the profit motive of private capitals” (Weber [1930] 2001: 17). However important this period may have been for the establishment of an “acquisitive manner of life” and a “crematist lifestyle” (ibid: 34), it was not sparkling with new ideas and innovations that would have triggered an avalanche of novelties. It rather imposed upon society a “worldly Protestant asceticism” and “acted powerfully against the spontaneous enjoyment of possessions; 19

Heinz D. Kurz

it constricted consumption, especially of luxuries” (ibid: 115). One idea, in particular, took possession of peoples’ minds: the firm belief that labour and industry was their duty towards God, and this duty requested the “accumulation of capital through ascetic compulsion to save” (ibid: 116). The period under consideration is characterised by what Marx had called the “production of absolute surplus value”: profits were increased first and foremost via an increase in the length of the working day and the intensity of labour and a lowering of wages, not via the introduction of new labour saving methods of production.

1.6.3

A “culture” of innovation

The Industrial Revolution was only possible because of a number of frequently disjointed institutional changes that took place after the Reformation and engendered the “Industrial Enlightenment” (Mokyr 2017). Thomas Robert Malthus’ view that mankind was doomed to live in misery and deprivation, was gradually replaced by a deep-rooted belief in human and social progress. A Baconian programme re-directed the agenda of natural philosophy towards solving practical problems that improved peoples’ living conditions. The establishment of learned societies and academies facilitated the cooperation of people who know things (propositional knowledge) and people who make things (procedural knowledge). The separation of science from metaphysics and especially a significant reduction of access costs to information and knowledge allowed European economies, in particular England and the Netherlands, to embark on a path of sustained economic growth, which eventually led to the “great divergence” between Europe and the rest of the world. Cultural entrepreneurs, including Newton, Galilei, Leibniz, and Spinoza, paved the way for economic entrepreneurs by inducing a cultural and political climate that was conducive to curiosity, experimentation, invention, and innovation. This culture was not the result of consciously planned and coordinated activities and reforms with an overarching goal in mind. It was rather the unintended consequence of a multiplicity of efforts and reforms in Europe. Adam Ferguson famously coined the phrase that social structures of all kinds are “the result of human action, but not the execution of any human design”. As a consequence, the flow of new, economically useful ideas gained momentum and set in motion the Industrial Revolution, during which both the variety of goods and of techniques to produce them began to expand swiftly. Due to a rising labour productivity, a given (and even a moderately rising) real wage bundle could be produced with less and less direct and indirect labour, and for a given length of the working day, a growing part of the labour performed became “surplus labour” and then profits. This is what Marx ([1867] 1954: 299) called “the production of relative surplus value”. It is the time when Schumpeterian entrepreneurs assume a central position in society and transform a swelling flow of inventions into profitable innovations, which revolutionise the way, how people work, live, and think. New ideas that materialise in innovations have particular consequences, that is, lead to particular facts. But these consequences and facts in turn induce new ideas, and so on. Ideas and facts, Schumpeter did not tire of stressing, co-evolve.

1.6.4 Schumpeter vs. Marx Apart from the phases of the rise and eventual endogenous erosion of ascetic Protestant ethic (Kurz 2021), from an evolutionary point of view the age of the production of absolute surplus value was much less interesting to Marx and Schumpeter than the innovation-driven 20

Joseph A. Schumpeter

age of the production of relative surplus value. But was Marx’s respective analysis fully compelling or did it suffer from a lack of a proper concept of the fulcrum of the novel socioeconomic dynamism – the agent of change, its demiurge? Schumpeter accused Marx that he, like before him Adam Smith, knew essentially only the capitalist, an agent possessed of money and wealth, but short on ideas. There was no cogent concept of the “entrepreneur”, who, while bursting with ideas, lacks the money to realise them and therefore has to turn to the capitalist and to banks to provide him with funds. Even worse, by confounding capitalists and entrepreneurs Smith and Marx were misled in assessing the sources of social tensions and class conflicts. Since such tensions were at the bottom of socio-economic change and development, a proper understanding of their causes and the way in which they discharged themselves was indispensable in a theory of the evolutionary (and occasionally revolutionary) dynamics of society. Which ideas clashed in this context, which erroneous or “fake news” dominated the debate about “Socialism vs. Capitalism”? Before we turn to this question, the following observation appears to be in place. Despite Marx’s blindness with regard to the outstanding role of the entrepreneur in the period of time under consideration, according to Schumpeter, Marx deserves to be praised for his remarkable achievements. “Nowhere”, Schumpeter stresses, “did he betray positive science to metaphysics” ([1942] 2008: 10). His analytical sociological perspective made him see through “the random irregularities of the surface down to the grandiose logic of things historical”. He adds: “And the outcome of his attempt to formulate that logic, the so-called Economic Interpretation of History, is doubtless one of the greatest individual achievements of sociology to this day” (ibid). Marx’s focus of attention on the mode of production and the relations of production as the main determinants of social structure and its inherent logic were “invaluable” working hypotheses that contained a large measure of truth. However, Marx is said to have totally underestimated the importance of “ideas and values” and of their bearers and therefore arrived only at partial truths.

1.7 1.7.1

Social conflicts – A flywheel of evolution Confounding capitalists and entrepreneurs

According to Marx, the main conflict in capitalist society was that between capitalists and workers. By wrongly subsuming entrepreneurs under capitalists, he was led to postulate also a conflict between entrepreneurs and workers. It deprived Marx’s in many respects illuminating analysis of much of its validity and, used as a weapon on the political battlefield, severely misled people and made them rally behind an ideology that was unsustainable. According to Schumpeter, there is no antagonistic relationship between entrepreneursinnovators, on the one hand, and workers, on the other. Thanks to entrepreneurs’ new combinations, productivity in the economy will grow, a higher profitability will accelerate capital accumulation and increase the demand for labour, which will in turn bid up real wages. Therefore, workers can on average be expected to benefit rather than suffer from entrepreneurial activity. To Schumpeter, both workers and entrepreneurs “are typical enemies of the given distribution of property as regards available goods. In many cases, both gain and lose together. Entrepreneurs are the best customers of workers. Any improvement in the conditions of workers starts from them” (Schumpeter 1912: 533). This is Schumpeter’s version of the doctrine of the unintended consequences of self-seeking

21

Heinz D. Kurz

human actions, championed by authors of the Scottish Enlightenment. Marx also missed the fact that while capitalists actually form a class, entrepreneurs do not: The entrepreneur employs his personality and nothing else than his personality. His position as entrepreneur is tied to his accomplishment and does not survive his vigour. It is essentially only temporary and cannot be bequeathed. The social position slips away from his follower that has not inherited with the prey also the lion’s claw. (1912: 529)

1.7.2

The dynamic performance of capitalism and socialism

Marx wrongly thought that consciously planned and administered socialism is dynamically superior to anarchic capitalism. How is this possible, asks Schumpeter, if the agent of change is absent? The misconception, Schumpeter surmises, must be the result of false economic doctrines clouding peoples’ minds. Besides socialist authors, one of the culprits in his story is John Maynard Keynes, who with his General Theory ([1936] 1972) is taken to have succeeded in pushing the cart of economics on to the wrong track. In Schumpeter’s view, economic crises and cycles do not reflect a recurrent breakdown or malfunctioning of the capitalist engine, but rather the way the system works. In Business Cycles, he emphasises: “Cycles are not, like tonsils, separable things that might be treated by themselves, but are, like the beat of the heart, of the essence of the organism that displays them” (Schumpeter 1939: v). Similarly, the idea that capitalist development could be rendered smooth and steady by means of a policy of aggregate effective demand is illusory. Capitalism, Schumpeter is convinced, is endangered by the claim that it is unjust and exploitative.

1.7.3 Income distribution and social justice This claim is supported with reference to a highly unequal distribution of the social product between different strata of society. Schumpeter responds to this in the following way. The “evolutionary character” of capitalism is closely related to a system of incentives that works with extraordinary rewards and punishments: Spectacular prizes much greater than would have been necessary to call forth the particular effort are thrown to a small minority of winners, thus propelling much more efficaciously than a more equal and more ‘just’ distribution would, the activity of that large majority of businessmen who receive in return very modest compensation of nothing or less than nothing, and yet do their utmost because they have the big prizes before their eyes and overrate their chances of doing equally well. Similarly, the threats are addressed to incompetence. ([1942] 2008: 73–74) By rewarding dexterity and innovativeness and punishing laziness and sloth, capitalism performs so well economically. Those who criticise it for its system of incentives apparently do not see that this is the source of its dynamic efficiency: one cannot have the latter without the former.9 The fact that the development is not smooth and steady, but comes in leaps and bounds, booms and busts, phases of overheating of the economy, followed by recessions or even depressions and mass unemployment, implies a serious problem for the reputation of capitalism. While those, who at a given moment of time suffer from the development, tend 22

Joseph A. Schumpeter

to blame it on the system, those who will eventually benefit from it, do not yet raise their voices. There is also a widespread misunderstanding of the role of successful big companies in all this. They are typically not seen as powerful engines of innovation and economic growth, earning monopoly rents because of their technological superiority, but as especially cunning institutions exploiting the public. A lack of a proper understanding of capitalism’s modus operandi is a dire source of the dwindling support of it.

1.7.4 Stabilisation policy The question cannot, of course, be supressed: is economic policy in Schumpeter’s view totally barren or can it somewhat fight the problems mentioned, as Keynes, for instance, argued? Can it mitigate the problem of unemployment and can it render income distribution less unequal without suffocating innovation and economic dynamism? As regards employment and economic activity as a whole, Schumpeter on the one hand sees clearly that in cases in which technological progress is highly disruptive and accompanied by sustained mass employment, a policy of stabilising aggregate effective demand and of supporting workers on the dole and their families is indispensable in order not to endanger the stability of the social order (Kurz 2015, 2016a, 2016b). On the other hand, he appears to think that in normal states of affairs, the economic system is able to absorb technological progress relatively smoothly if left to its own devices. His interpretation of the Great Depression in the Business Cycles (1939: chap. XIV) is interesting in this respect: he traces it back to the unfortunate concurrence of troughs of three types of economic cycles: Kitchins (inventory cycles), Juglars (business cycles), and Kondratieffs (long waves). However, he apparently regards the probability of a repetition of such a concurrence as negligible. Schumpeter therefore appears to feel justified to rely essentially on the so-called self-healing forces the socio-economic system activates from within in case of dangers threatening its existence – an expression of a crucial self-preserving evolutionary mechanism. Actually, he goes a step further and interprets the private and public sectors – the economy and the state, for short – as forming two interacting parts of an intimately intertwined whole. Hence the system is able to preserve its basic structure, if the two spheres interact in a mutually beneficial way. Fettering the private sphere will in the long run turn out to be detrimental also to the public sphere and move the system to the brink of socialism.10 In the History of Economic Analysis (1954: 1173 fn. 3), Schumpeter insists that a progressive taxation of profits is not an exogenous reason of the weakening of economic dynamism, as is frequently maintained in the literature. It is rather a genuine element of the jointly evolving capitalist economy and the capitalist state. It makes no difference, he emphasises, whether in a “profit economy” profitable investment opportunities dry up, as Keynes and Alvin Hansen argued in their stagnation theories, or the realised profits are taxed away. Schumpeter’s fierce opposition to Keynes’ aggregate effective demand policy is rooted in the following considerations. While in the short run, such a policy might on the surface look like a proper means to salvage capitalism and avoid the march into socialism, in the medium and long run it has exactly the opposite effects. It causes a weakening of the innovationgenerating engine and thereby renders capitalism feeble and vulnerable. While the British economist also wished to defend capitalism against its enemies (opting, of course, for significant changes in its institutional framework and people’s motivations and lifestyles), his basically static economic analysis made him suggest the wrong means to do this. A dynamic, evolutionary analysis was badly needed, and Schumpeter was convinced to have elaborated it. 23

Heinz D. Kurz

1.7.5

Income distribution policy

Schumpeter stresses that the capitalist incentive system works with excessive punishments and rewards. Hence, in his view less extreme schemes would do to call forth the desired innovative activity. In his Theory of Economic Development, he also stressed that innovations could be financed other than by bank credit (e.g., by means of venture capital) and would thereby also reduce the fundamental instability of the banking system, which deepened and prolonged economic crises. However, he largely left it at that. Apparently, he saw this justified in terms of the following considerations. First, an optimal incentive system could only be defined with respect to the particular features of an innovation and would have to be changed with them. But would it be practicable to elaborate a mechanism design tailored to the specificities of waves of technological change, whose properties become clear only in the course of time? Secondly, Schumpeter is convinced that even glaring inequalities of income and cases of “injustice” weigh relatively little, if seen in the proper perspective. Whilst it is true that the commercial society generates “unrestricted inequalities”, it is also true that workers in it are paid significantly higher wages than “the equal incomes” paid “in egalitarian socialism” (Schumpeter [1942] 2008: 191). This is ignored by the critics of capitalism and fuels the hostility towards it, thereby refusing “on principle to take account of the requirements of the capitalist engine”, which becomes “a serious impediment to its functioning” (ibid: 154). Schumpeter’s conclusion is straightforward: economically, capitalism is superior to socialism, but this fact is not well understood even by its most prominent supporters. But what can be said about other important features of the two different social orders – their cultural achievements and the right of people to have a say in political and social affairs?

1.8

Culture and democracy

Does Schumpeter’s praise of capitalism carry over from the economic to the cultural and political sphere? Capitalism, he observes, generates “not only modern technology and economic organization, but all the features and achievements of modern civilization are, directly or indirectly, the products of the capitalist process”. He adds: “They must be included in any balance sheet of it and in any verdict about its deeds or misdeeds” ([1942] 2008: 125).11

1.8.1

Culture

While with regard to culture, his account is again in favour of capitalism, he is far from being enthusiastic.12 Capitalism, he insists, shapes mental habits, rationalises modes of thought and behaviour, submits everything to a utility calculus, and, echoing Adam Smith in Book V of The Wealth of Nations (Smith 1976), causes an “anti-heroic” and “pacifist” attitude. Capitalism liberates people from earlier constraints and compulsions, but subjugates them to new ones. Schumpeter appears to share elements of Max Weber’s cultural pessimism. He points out that spreading utilitarianism leads to the total annihilation of the bourgeois family and even of all ideologies. Capitalism does not only produce a growing variety of commodities in growing amounts, it also “turns out” a particular “kind of human beings … that it then leaves to their own devices, free to make a mess of their lives” (ibid: 129). There is no salvation to be expected from the world created by capitalism (nor, of course, from that created by socialism).

24

Joseph A. Schumpeter

1.8.2 Democracy In Part IV of Capitalism, Socialism and Democracy, Schumpeter is sceptical that in socialism democratic conditions can prevail. But his view of the role of democracy in capitalism is also all but exhilarating. Democracy he defines as a “political method” to reach collective decisions. This definition allows for different relationships between democracy and liberty and is neither based on the utilitarian interpretation of democracy in the tradition of Jeremy Bentham nor on Jean-Jacques Rousseau’s social contract theory, both of which Schumpeter rejects. The essence of politics is a “competitive struggle for power and office” (ibid: 282) and not about who promotes better the public good. Echoing the view of his former teacher Friedrich von Wieser, this involves “attempts to contact the subconscious” (ibid: 263) and capture the minds of people. Not the people decide their own destiny, but politicians craving for recognition and dominance do. The struggle for power is wasteful and exhausting and impairs the effectiveness of this form of government. Schumpeter’s view of democracy is sober, occasionally cynical, and can hardly be called enthusiastic. While modern democracy evolved alongside capitalism, it is in principle also compatible with socialism. Socialism in fact ideally implies the extension of democracy to the economic sphere. This may and typically will, however, have negative implications for economic efficiency and innovation. Since under socialism, contrary to capitalism with its private economic sector, there is no separation of powers, it is much easier for politicians to seize full control over society. Schumpeter concludes his respective reasoning with the words: “As a matter of practical necessity, socialist democracy may eventually turn out to be more of a sham than capitalist democracy ever was” (ibid: 302).

1.9

Concluding observations

Can Schumpeter be considered one of the founders of evolutionary economics? The answer given here is a resounding yes, but the evolutionist character of his work is not the result of a simple transfer and application of concepts of evolutionary biology to the social sciences. In fact, Schumpeter was convinced that the subject matter of the latter defied such a transfer. However, he saw economy, society, culture, and politics as closely intertwined parts of an evolving system that could only be investigated in terms of a theory of evolutionary dynamics. The system to be understood was a self-organising socio-economiccultural-political entity possessed of its own internal logic, generating incessantly change from within and inducing processes of self-transformation. Schumpeter concluded that the social sciences had to elaborate their own concept of evolution: they could derive inspiration from Darwin and evolutionary biology, but what was needed was conceptual and analytical innovation, not imitation. We owe Schumpeter important insights and steps forward in this regard. Luckily enough, he did not have to start from scratch, because other social scientists had already begun to pave the way towards evolutionary economics. While Schumpeter strictly rejected the idea that economists could learn much from evolutionary biology, we may nevertheless use some of its ideas and concepts to paraphrase his point of view. The concept of random variations suggests itself with view to new generations of people possessed of what we might, for short, call entrepreneurial genes. However, whereas in Darwin’s view, the environment selects those individuals possessed of the most advantageous properties vis-à-vis the existing environment, in Schumpeter’s view entrepreneurs’

25

Heinz D. Kurz

outstanding feature is their capacity to overthrow the conditions they face instead of fitting themselves to them. They seek to do away with the perceived constraints, leave the trodden path, establish new businesses and work routines, introduce new rules of the game, and erect a business empire and possibly a dynasty. New products and methods of production may also be compared to mutations. They share with them their path-dependency and play a crucial role in the battle of survival imposed by competitive conditions. According to Schumpeter, there are the following differences between the concept of evolution in biology and in the social sciences. First, human actions are characterised by intentionality: agents seek to anticipate and estimate the consequences of their actions and may swiftly change their behaviour, if necessary or preferable. In this regard, they are assisted by their capacity to learn and by their creativity. There is no equivalence to this in biology. Schumpeter’s thinking about socio-economic development is therefore not rooted in the natural sciences, but rather in a theory of human culture. The second one concerns Darwin’s idea of a fairly steady, gradual, and slow adjustment process in biological evolution. Economic change, Schumpeter insists, is instead characterised by discrete innovations that are frequently disruptive and cause change in a relatively short period of time. Capitalism proceeds in terms of a process of “creative destruction”, its development is not characterised by smooth transitions between subsequent states of the socio-economic system. Capitalism is an evolutionary system characterised by an irresistible urge to creative destruction from within. It can never be stationary, but is endogenously forced to develop and evolve. Its overall balance sheet notes important achievements, especially as regards the quantity and quality of goods it turns out, but also serious failures – and it is a system that nourishes forces that threaten its survival.

Notes 1 I am most grateful to Kurt Dopfer for valuable comments and suggestions on an earlier version of this paper. It goes without saying that all remaining misconceptions are entirely my responsibility. 2 For summary accounts of Schumpeter’s works, see, for example, McCraw (2007) or Kurz and Sturn (2012). 3 While Schumpeter did not particularly admire Adam Smith’s work, it is only fair to point out that it contains several evolutionist elements ante litteram ( Coase 1976). These include an early expression of the principle of circular and cumulative causation in the doctrine of the social division of labour, which revolves around dynamically increasing returns ( Young 1928). But Smith used the theoretical concepts and tools he had forged also in order to explain, first, the transition from a feudal to a “commercial society” and, secondly, the endogenous factors building up that threaten the continuation of the process of civilisation in the long run. There is a parallel between these themes and Schumpeter’s emphasis on the innovative capacity of capitalism, on the one hand, and its replacement by socialism, on the other. Interestingly, Smith argued that feudalism falls victim to the introduction of new methods of production and new luxury goods produced in manufactures. These goods “furnished the great proprietors with something for which they could exchange the whole surplus produce of their lands, … and thus, for the gratification of the most childish, the meanest and the most sordid of all vanities, they gradually bartered their whole power and authority” (WN III.iv.10). In this way, “A revolution of the greatest importance to the public happiness, was … brought about by two different orders of people, who had not the least intention to serve the public” (WN III.iv.17). The economic success of the commercial society had, however, a serious drawback: it gradually replaced a “martial” by a “commercial spirit” and established the “principle of avarice”. Acquiring great wealth in business became an attractive alternative to the offspring of the upper classes of society, to whom in a rude society “nothing is honourable but war”. It therefore became inconvenient for the rich and educated to do military service. As a consequence, “it fell to the meanest to defend the state”, which, however, were hardly able to

26

Joseph A. Schumpeter

4

5

6 7

8 9 10 11 12

understand commands and operate sophisticated weaponry. A decline of soldierly virtues and fitness for war exposed civilised societies to envious and warlike barbarous neighbours. This endangered severely the “system of natural liberty” and the continuation of the process of civilisation. Smith put all hope in the “wisdom of the state” (WN V.i.a.14), which is supposed to ward off the threat in terms of a standing army and a militia that keeps the martial spirit alive, for “defence is of much more importance than opulence” (WN IV.v.a.36). The following observations are appropriate here. As is well known, Marshall ([1890] 1961: 772) insisted that economics “is a branch of biology broadly interpreted” and that therefore “the Mecca of the economist lies in economic biology rather than in economic dynamics”. It might therefore come as a surprise that he nevertheless endorsed the mechanical perspective, which revolves around the concepts of static equilibrium and optimal economic states. He justified his choice in terms of the much greater complexity of biological conceptions compared to mechanical ones (ibid). There is, however, also the fact that Marshall was essentially a Spencerian and not a Darwinian and believed in the “optimising” nature of evolutionary processes from the not so good to the better ( Hodgson 1993). Such an outlook can easily be accommodated within a marginalist framework. For an entirely different evolutionist perspective, see Veblen (1898). Schumpeter stressed that both capitalism and socialism are, each within its own limits, “culturally undetermined”, that is, there are several forms of each of them. In the final part of Capitalism, Socialism and Democracy ([1942] 2008), he discusses in some depth different kinds of socialism, but for the most part, his attention focuses on the Soviet Union under Stalin, which had established itself as the leading socialist power in the world. In the following, we put on one side the problem of different variants of the two types of social order. For the following, see also Kurz (2018a: 14–17). As is well known, Schumpeter, following in the footsteps of Marx, conceives of ideology not only as an obstacle in the way of scientific progress, but also as an engine in the elaboration of theories. “Pre-analytic visions” may give rise to new questions and hypotheses and spur scientific inquiry. He insists that there is simply no social theory that is totally ideology-free. Marx was wrong in assuming that his own theory was. Readers interested in the fascinating theme are, of course, asked to consult the works cited or some of the contributions dealing with aspects of them. See, for example, Gehrke (2019) and Kurz (2018b, 2020). Different styles or cultural variants of capitalism do, of course, differ also with regard to the character and magnitude of the incentives they allow for and of the taxation of gains and treatment of losses. Schumpeter is surprisingly much less concerned about an insufficient fostering of the public sector. For the following, see also the detailed account in Kurz (2020). Interestingly, Schumpeter (1954: 444 n. 18) gets close to insinuating that the great Darwin was a product of the civilisation of capitalism, which may be read as speaking strongly in favour of the cultural fecundity of it.

References Coase, R. H. (1976). Adam Smith’s View of Man. Journal of Law and Economics, 19: 529–546. Darwin, C. (1859). On the Origin of Species by Means of Natural Selection. London: John Murray. Darwin, C. (1871). The Descent of Man, and Selection in Relation to Sex. London: John Murray. Dopfer, K. (ed.) (2001). Evolutionary Economics. Boston, Dordrecht, and London: Kluwer Academic Publishers. Dopfer, K. (ed.) (2005). The Evolutionary Foundations of Economics. Cambridge: Cambridge University Press. Dopfer, K. (2016). Evolutionary Economics. In G. Faccarello and H. D. Kurz (eds.), Handbook on the History of Economic Analysis, vol. III: Developments in Major Fields of Economics. Cheltenham and Northampton: Edward Elgar, pp. 175–193. Dopfer, K. and Nelson, R. R. (2018). Evolution of Evolutionary Economics. In R. R. Nelson (ed.), Modern Evolutionary Economics. An Overview. Cambridge: Cambridge University Press, pp. 208–229.

27

Heinz D. Kurz Dopfer, K. and Potts, J. (2008). The General Theory of Economic Evolution. London: Routledge. Gehrke, C. (2019). Joel Mokyr’s A Culture of Growth: A Book Roundtable. The European Journal of the History of Economic Thought, 25(6): 1493–1536. With contributions by Christian Gehrke, Erik Buyst, Heinz D. Kurz, Richard Sturn, and Joel Mokyr. Hodgson, G. M. (1993). The Mecca of Alfred Marshall. Economic Journal, 103: 406–415. Keynes, J. M. ([1936] 1972). The General Theory of Employment, Interest and Money. London: Macmillan. Reprinted in The Collected Writings of John Maynard Keynes, Vol. VII: London: Macmillan. Kurz, H. D. (2012). Schumpeter’s New Combinations. Revisiting His Theorie der wirtschaftlichen Entwicklung on the Occasion of Its Centenary. Journal of Evolutionary Economics, 22: 871–899. Kurz, H. D. (2015). The Beat of the Economic Heart. Joseph Schumpeter and Arthur Spiethoff on Business Cycles. Journal of Evolutionary Economics, 25: 147–162. Kurz, H. D. (2016a). Schumpeter und Keynes. Gemeinsam gegen den Strom, getrennt zu neuen Ufern. In H. Hagemann and J. Kromphardt (eds.), Keynes, Schumpeter und die Zukunft der entwickelten kapitalistischen Volkswirtschaften. Marburg: Metropolis, pp. 109–137. Kurz, H. D. (2016b). Is there a “Ricardian Vice”? And What Is Its Relationship with Economic Policy Ad”vice”? Journal of Evolutionary Economics, 27: 91–114. Kurz, H. D. (2018a). Making His Own Sense of Marx. Schumpeter’s Adoption-cum-Adaptation of Marxian Ideas. Marx-Engels-Jahrbuch, 2017(1): 80–102. Berlin: DeGruyter. Kurz, H. D. (2018b). Hin zu Marx und über ihn hinaus: Zum 200. Geburtstag eines deutschen politischen Ökonomen von historischem Rang. Perspektiven der Wirtschaftspolitik, 19/3: 245–265. Kurz, H. D. (2020). Kapitalismus, Sozialismus und Demokratie: Schumpeter’s Entwurf einer histoire raisonnée der Moderne. In J. A. Schumpeter (ed.), Kapitalismus, Sozialismus und Demokratie, 10th complete German edition of a translation of Schumpeter ([1942] 2008). Tübingen: Narr Francke Attempto, pp. XI–LXIV. Kurz, H. D. (2021). Max Weber on the “Spirit of Capitalism”. Economic Growth and Development in the Antechamber of the Industrial Revolution. Investigación Económica, 80(318): 32–71. Kurz, H. D. and Sturn, R. (2012). Schumpeter für Jedermann. Von der Rastlosigkeit des Kapitalismus. Frankfurt am Main: Frankfurter Allgemeine Buch. Lorenz, H.-W. (2020). Der Darwinismus in der Nationalökonomik. In P. Spahn (ed.), Von Marx & Engels zu Nelson & Winter (und darüber hinaus). Berlin: Duncker & Humblot, pp. 1–53. Marshall, A. ([1890] 1961). Principles of Economics. 8th ed. Variorum Edition. Two vols. London: Macmillan. Marx, K. ([1867] 1954). Capital, Volume I. Originally published in German in 1867, English translation. London: Lawrence and Wishart. Mayr, E. (1982). The Growth of Biological Thought. Diversity, Evolution, and Inheritance. Cambridge, MA: Harvard University Press. McCraw, T. (2007). Prophet of Innovation: Joseph Schumpeter and Creative Destruction. Cambridge, MA, and London: Belknap Press of Harvard University Press. Metcalfe, J. S. (1998). Evolutionary Economics and Creative Destruction. In The Graz Schumpeter Lectures. London: Routledge. Metcalfe, J. S. (2005). Evolutionary Concepts in Relation to Evolutionary Economics. In K. Dopfer (2005), S. 391–430. Mokyr, J. (1991). Evolutionary Biology, Technological Change, and Economic History. Bulletin of Economic Research, 43(2): 127–149. Reprinted in G. M. Hodgson (ed.) Economics and Biology (Brookfield, VT and Aldershot UK: Edward Elgar, 1993), pp. 471–93. Mokyr, J. (2017). A Culture of Growth: The Origins of the Modern Economy. In The Graz Schumpeter Lectures. Princeton, NJ: Princeton University Press. Nelson, R. (ed.) (2018). Modern Evolutionary Economics. An Overview. Cambridge: Cambridge University Press. Nelson, R. R. and Winter, S. G. (1982a). An Evolutionary Theory of Economic Change. Cambridge, MA: Harvard University Press. Nelson, R. R. and Winter, S. G. (1982b). The Schumpeterian Tradeoff Revisited. American Economic Review, 72(1), S: 114–132. Nelson, R. R. and Winter, S. G. (2002): Evolutionary Theorizing in Economics. Journal of Economic Perspectives, 16(2): 23–46.

28

Joseph A. Schumpeter Ruse, M. (1988). Molecules to Men: Evolutionary Biology and Thoughts of Progress. In M. H. H. Nitecki (ed.), Evolutionary Progress. Chicago, IL: Chicago University Press. Schumpeter, J. A. (1908). Das Wesen und der Hauptinhalt der theoretischen Nationalökonomie. Berlin: Duncker & Humblot. Schumpeter, J. A. (1912). Theorie der wirtschaftlichen Entwicklung, 4th ed. 1934, Berlin: Duncker & Humblot. Schumpeter, J. A. ([1932] 2005). Development. Journal of Economic Literature, 43(1): 108–120. Translation of an unpublished manuscript entitled “Entwicklung”, Bonn 1932. For the German version, see: www.schumpeter.info/schriften/entwicklung.htm. Schumpeter, J. A. (1939). Business Cycles, 2 vols. New York, NY: McGraw-Hill. Schumpeter, J. A. ([1942] 2008). Capitalism, Socialism and Democracy. Reprint of third ed., published in 1950, as First Harper Perennial Modern Thought Edition. New York, NY: Harper & Row. Schumpeter, J. A. (1954). History of Economic Analysis. Edited by E. B. Schumpeter, Oxford: Oxford University Press. Smith, A. ([1776] 1976). An Inquiry into the Nature and Causes of the Wealth of Nations, two vols. In R. H. Campbell and A. S. Skinner (eds.), The Glasgow Edition of the Works and Correspondence of Adam Smith. Oxford: Oxford University Press. In the text referred to as WN. Veblen, T. (1898). Why is Economics Not an Evolutionary Science? Quarterly Journal of Economics, 12(4): 373–397. Weaver, R. M. (1948). Ideas have Consequences. Chicago, IL: University of Chicago Press. Weber, M. ([1930] 2001). The Protestant Ethic and the Spirit of Capitalism, translated by T. Parsons. Reprint. London: Routledge. Young, A. (1928). Increasing Returns and Economic Progress. Economic Journal, 38(152): 527–542.

29

2 THORSTEIN BUNDE VEBLEN A founder of evolutionary economics Helge Peukert

2.1

Introduction

Thorstein Bunde Veblen (1857–1929) was the 6th of 12 children of Norwegian immigrants. He grew up on a farm in Nerstrand, Minnesota. Alongside John Commons, he is the founder of critical institutionalism. In view of his economic writings (1898, 1904, 1919a, and 1923, compiled in Veblen 1994), he is also regarded as the founder of evolutionary economics. His contributions indicate strong social concerns and sympathy for the common man. Veblen was a polymath whose dissertation was about Kant’s and Peirce’s philosophy. He developed a motivational theory (1914) and wrote methodological (1919a) and clairvoyant political writings (1915; for the overall reception of Veblen, see Wood (Ed.) 1993). Veblen’s school of thought persists in the Association for Evolutionary Economics (AFEE) and in the Journal of Economic Issues (JEI) published by AFEE (on his biography as a “Victorian firebrand” see Jorgensen and Jorgensen 1999).

2.2 Theoretical approach and evolutionary epistemology In his early article, Why is economics not an evolutionary science? (1898), Veblen stood out as a classical contributor to an evolutionary-institutional research project (Pyka and Dopfer (Eds.) 2019, Vol. 1, 1). It is directed against mechanical-physical, a-historical, and utilitarianhedonistic approaches, which began to prevail in economics. “The hedonistic conception of man is that of a lightning calculus of pleasures and pains, who oscillates like a homogenous globule of happiness under the impulse of stimuli” (Veblen 1898, 393). For Veblen, the endogenous change of economy and society, both in constant evolution and change, was the explanandum. According to the prevailing interpretation, Veblen made an unambiguous attempt to construct evolutionary economics. He saw instincts, habits, and institutions in economic evolution as analogous to genes in biology. For Hodgson, for example, Veblen is an important – although unsystematic – founder of evolutionary economics (1993, 123–138; see also Dopfer 2000): Veblen applied a universalized Darwinian natural selection principle and the transformation of individuals and institutions as an ongoing process in combination with habit psychology.

30

DOI: 10.4324/9780429398971-4

Thorstein Bunde Veblen

Habits precede rational deliberation. But Hodgson holds that for Veblen, individual human agency and social institutions are nevertheless mutually constituive. But there exists at least a complementary dimension in Veblen’s thinking. It is already evidently manifest in his early article on evolutionary economics: It will not even hold true that our elders overlooked the presence of cause and effect in formulating their theories and reducing their datat to a body of knowledge … The difference [between older and evolutionary concepts] is a difference of spiritual attitude or point of view … it is a difference in the basis of valuation of the facts for the scientific purpose, or in the interest from which the facts are appreciated. (1898, 377) For Veblen, every theory depends on preconceptions or paradigms which impute coherence to the facts dealt with (for evidence see Peukert 2001). He comprehensively received the multifaceted discourse universe of his time, including Kant’s and Peirce’s epistemologies, which inspired him to design a full-fledged implicit evolutionary epistemology (see the compilation by Camic and Hodgson (Eds.) 2011). Kant’s message for him was that knowledge always depends on preanalytic schemes of apprehension. But also Peirce’s objective idealism and consensus theory of truth strongly influenced Veblen. According to Peirce, matter is a frozen spirit that develops habitual attitudes on the basis of never-ending creativity and always has to choose voluntaristically among innumerable alternatives. The main function of thinking and scientific research is not to reveal “truth”, but to bring the mind conservatively to rest with more or less provisional preconceptions. Such reifications are necessary to establish some order. From Veblen’s (1919a) evolutionary-deconstructive point of view in the tradition of Peirce, all worldviews are ultimately based on a certain self-deception. This applies both to the dramatic cosmologies of the early “good” hunters and gatherers, who believed in animisticteleological projections, and to the metaphysical-hierarchical concepts of early barbarism (feudalism), which contained an ideological justification for authoritarian rule and servitude. In modern capitalism with its integrated machine system, one believes in the law of clear causes and causally attributable effects and the assumption of normal equilibrium states. Since the beginning of the 20th century, the last phase has been a worldview of evolutionary change with impersonal, cumulative, not clearly causally attributable processes. They also lead to a “self-contamination” of the mind through a supposed lack of direction and impotence towards the natural drift of social development. It implies a nihilistic moral and aesthetic indifference. This worldview stigmatizes questions of “why” and “where” as naïve, which is often (inconsistently) combined with metaphysical assumptions about a welcome evolutionary progress (e.g. thanks to competition as a discovery process). Interpreting Veblen only as an evolutionary economist would escape the peculiarities of his approach. But in the summarizing part five below the nevertheless strong correspondence between Veblen’s institutionalism and evolutionary approaches will be highlighted. As a quasi-existentialist constructivist, enlightened by Kant and Peirce, he distrusted all reifying, non-evolving worldviews and concepts. In the theory of the leisure class, therefore, often floating-ambiguous statements occur that are supposed to bring the reader into a skeptical mode and of the non-fixity of mind and matter. With his ironic-satirical style, Veblen acknowledges the inescapable epistemological problem of self-referentiality (Dopfer 2000, 99). When every proposition is considered relative: Does this also refer to this metastatement itself? 31

Helge Peukert

2.3 Applied evolutionary economics: The theory of the leisure class In 14 chapters, on around 350 pages, Veblen published his main work, which is considered a classic of social criticism. Including the body of sociological, anthropological, philosophical, and economic knowledge, he analyzed the institutions and collective habits of his time. However, the study goes beyond the analysis of contemporary historical phenomena of American society. He offers an alternative social theory with an economic-sociological analysis of prestige based on the ability to afford wasteful consumption at the center. His book is easily underestimated because Veblen used amusing and ironic formulations. The satirical style reflects his opposing attitude and distance from the well-established, frozen habits of thought. Veblen’s central thesis is that – after the level of subsistence has been exceeded – striving for prestige is the driving force behind the development of human societies. It stimulates growth and competition and also shapes the state, which is a power apparatus in the interest of the wealthy. Most economic activities are aimed at waste and status demonstration, not at intrinsic material fulfillment of needs. Just as the yam tubers of the Big Men in New Guinea that rot behind their huts seem irrational and possibly ridiculous as a status symbol for us, so consumption and profit in today’s growth societies have the same motivational origin. In feudalism, the status demonstration was about leisure, in capitalism about goods consumption and emulative display. In every society that knows private property, the individual must have at least as much in the interest of his inner peace as that with which he puts himself on the same level; and it is extremely beneficial to have something more than others. But as soon as someone has acquired new goods and got used to the new property, it is no longer a pleasure for him than the old one. This is because there is always a tendency to view the current situation only as a starting point for further growth in goods. (2007/ 1899, 47) Veblen designs an alternative stage theory that contradicts social contract theories of e.g. Locke or Hobbes, which ultimately assume synergetic advantages for everyone. He contrasts them with a theory of inequality and violent superimposition. He differentiates the phase of “good” savagery (hunters and gatherers) with that of “bad” barbarism. The latter is divided into the original barbarism in feudalism, followed by that of the capitalist phase. Both barbaric phases are characterized by the existence of non-productive, parasitic, idle groups like priests, warriors, politicians, athletes, professors, and absentee owners. The shareholders are the new feudal lords. Their cunning managers are primarily interested in profit and the most expensive sales possible. Veblen introduces a fundamental dualism between productive (e.g. peasant work) and predatory activities in feudalism (robbery, warfare) versus industrial (real value-adding) and pecuniary (money-oriented) activities in capitalism. This distinction is embedded in his (deliberately ambiguous) suggestion of three constructive “instincts”: Productive workmanship, idle curiosity, and the parental bent (1914). They are often superimposed by predatory instincts in more developed societies. In the developed stage of capitalism, mass production dominates in oligopolistic, scientifically-based key industries. A cross-linked technostructure with unprecedented efficiency is created thanks to the technical intelligence of the engineers. The (small) trades and

32

Thorstein Bunde Veblen

agriculture are grouped around this core. A productive majority of the population is faced with a parasitic upper class, a highly questionable division of labor. The prevailing habits of thought are mostly determined by the upper class, which succeeds in devaluing productive work and allowing wealth to be regarded as honorable regardless of how it was generated. They exhibit a male-aggressive social character mirroring the hegemonic money-related “culture”, exemplified in Chapter 10 with North American sports and American football in particular. According to Veblen, private property did not arise through the legitimate appropriation of laboriously manufactured work. Originally, it came about through the robbery of women from other hordes as a trophy and for their exploitation as “workhorses”. Women were the first property titles for social distinction. Veblen contradicted early bourgeois interpretations which legitimize private property with righteous work as the foundation of the institution of property, as a motivation to save, etc. Property was later primarily a means of luxury consumption and to exploit laborious average men and women. He castigated widespread legal fictions with regard to the supposedly productive effects of property. For him, the legal and habitual system of his time was based on the principles of law and morality which were developed in the 16th, 17th, and 18th centuries, i.e. the traditional principles of personal self-regulation, free contracts, and equal opportunities. Securing and improving a desired standard of living as an indicator of an increase in material wellbeing is not a contradictory goal. It was considered the final objective of economic activity at least since Adam Smith’s The Wealth of Nations. Veblen opposes this view with an irrationality thesis directed against growth-oriented mainstream economics, but also against e.g. sociological rationalization hypotheses. “But since the struggle is primarily a race for reputation and respectability, both based on a discriminatory comparison, this goal [of increasing wellbeing by raising the material standards] can never be achieved” (2007/1899, 48). It is impossible for everyone to be better off than their neighbors. According to Veblen, competition is therefore a harmful process. Veblen consequently asks the question that the late Keynes afterwards raised: If it were really true … that the incentive to accumulate goods is only the concern for existence and the desire for material comfort, then it should also be possible to fully satisfy the economic needs of a society at a certain point in material development. (2007/1899, 48) Veblen transfixed everyday phenomena and behavior from a sarcastical and critical hermeneutical perspective in his magnum opus. He vividly addresses the display of highbrow knowledge of dead languages, good writing skills, house music, collecting valuable images, following arbitrary, often ceremonially sophisticated decency rules and conventions, keeping more or less exotic – preferably more expensive – servile (dogs) or faster animals (horses), hunting and the possession of firearms as a barbaric relic, exclusive furnishings and memberships in exclusive clubs and associations, fashionably exquisite clothing and jewelry, the celebration of lavish festivities, etc. In contrast to Marx, who thought that the lower classes would defend themselves against their oppressors and rebel, Veblen suspects that they are more inclined to imitate their masters as much as possible and to copy their lifestyle. Veblen’s approach has rightly been regarded as an early attempt at a critical evolutionary economic approach. But he is often underestimated as an economist. In connection with his 33

Helge Peukert

initially mentioned economic writings in the narrower sense, his theory of the leisure class also contains a fundamental criticism of still prevalent economic theorizing with his analysis of the bandwagon-, the snob-, and the Veblen effect. Besides the “more is better” they contradict essential assumptions of the mainstream (at least in the textbooks), which assumes that individual benefits and consumption are independent and do not influence each other; that one can get from the individual demands through aggregation to the total demand; that benefits are intrinsic (i.e. the product qualities and not external demonstration effects count) and he questions that a falling (rising) price leads to increasing (falling) demand (for details see Peukert 1998, Chapter 6). In any case, he molded critical old institutionalism. Beyond the details characteristic of Veblen, he founded – along with Commons and others – a research program, the main features of which Atkins summarized accurately as follows: (1) group behavior, not price, should be central in economic thinking; (2) more attention should be given to uniformities of custom, habit, and law as modes of organized economic life; (3) individuals are influenced by motives that cannot be quantitatively measured; (4) economic behavior is constantly changing; therefore, economic generalizations should specify limits of culture and time to which they apply; (5) it is the task of the economist to study the sources of the conflict of interests in the existing social structure as an integral factor rather than a something diverging from a hypothetical norm. (Atkins 1932, 111) These five general characteristics show that Veblen agrees with an evolutionary approach in general, but he held also contrasting views e.g. regarding New Institutional Economics.

2.4

Evolutionary heterodox supply-side economics: Technostructure, absentee owners, and speculative finance

Especially in The Theory of Business Enterprise (1904) and Absentee Ownership (1923), Veblen developed a contrarian theory to the prevailing (neoclassical) market process approaches and the law-like determination of scarcity inducing prices as a result of the supply and demand of powerless isolated agents. For Veblen, three action patterns interact as follows: The productive mechanical system of industry, the strongly acquisitive-predatory driven price system, and the national establishment (1919b), which is a predacious enterprise with the application of fraud and force too. For Veblen (1915), a capitalist government is a business government. Mass production, based on fossil fuels and scientification, characterizes the core of the U.S. economy. An interdependent, linked, and webbed technostructure with hitherto unequaled mechanical efficiency arose which tends to be a more or less perfect automatic mechanism. Around the key industries, small businesses and agriculture are grouped. The industrial process, e.g. the size of firms, the combination of inputs, etc., has primarily a mechanical nature which is in principle independent of value and price aspects. As a counterpoint to mainstream economics, Veblen’s view may be somewhat exaggerated. But his focus on intertwined technological relations is shared by most evolutionary economists: It is a comprehensive system of interdependent working parts, organised on a large scale and with exacting articulation of parts … , - works, mills, railways, shipping, 34

Thorstein Bunde Veblen

groups and lines of industrial establishments, all working together on a somewhat delicately balanced plan of mutual give and take. No one member or section of this system is a self-sufficient industrial enterprise. (1919b, 51–52) On the eve of Sraffa’s articles on returns, Veblen already put into question the assumptions of relatively soon decreasing returns in mainstream economics. In the key industries, usually falling marginal costs and increasing returns were the rule. The thesis of the interlocking technostructure is problematic for the mainstream theory of distribution (J.B. Clark) according to which the marginal product of the individual factors of production determine their shares. The state of the industrial art is a joint stock of knowledge derived from past experience, and is held and passed on as an indivisible possession of the community at large. It is the indispensable foundation of all productive industry, of course, but except for certain minute fragments covered by patent rights or trade secrets, the joint stock is no man’s individual property. For this reason, it has not been counted in as factors in production. (Veblen 1921, 28) In the influential study by Berle and Means (1932) on managerial capitalism, Veblen’s analysis and criticism of absentee owners and the large industrial machine system are linked (1923, see also Hudson (Ed.) 2016). “Capital” is the common societal stock of (technological) knowledge. In capitalism, private companies try to transform this common stock into tangible and appropriable technologies, goods, and services to generate individual profits and rents (which are often guaranteed as monopoly rights by the state). Land rents and (large-scale) landed property also play an important role in (re)distribution and as effortless income. Technological progress should be ascribed to inventors, natural scientists, engineers, and electricians and less to pecuniary-oriented entrepreneurs (Veblen 1921, 60–61). To repeat: Following the common knowledge stock perspective, the imputation of productive services of each factor of production does not make sense and increasing returns would preclude a “just” return for labor in a marginalist perspective. Increasing returns exhibit a natural tendency for concentration and the exercise of oligopolistic power and political pressure. This should have far-reaching consequences for antitrust policy. The interdependent machine system also contradicts the assumption of the non-interdependence of the production processes. It implies a flak of e.g. the Walrasian pure exchange approach with separate stocks of goods as parameters. Veblen developed an antithetical theory of the firm and especially of limited liability as a historically specific institution, combining it with the financial aspects. In his view, a decoupling of the ownership of the means of production and the practical guidance especially in the core industries took place. The shareholders as loss-of-function absentee owners became the new feudal lords. In contrast to the earlier constructive role of the owners as initial investors, inventors, and organizers of the production processes, the absentees exploit the increasing scale of the industrial plants and their operations by administrative control as did the landlords in times of political dominion. They are the principles the managers are their agents. “The business man’s place in the economy of nature is to ‘make money’, not to produce goods” (1919b, 92). 35

Helge Peukert

In contrast to theories of manager capitalism with managers as the new independent ruling class, Veblen sketched a shareholder value model and he saw their role rather as string puppets in assistance of the absentees. Managers pursue inter alia a sabotage strategy for profit maximization: They maintain profitable prices by limiting output, monopolizing patents and copyrights, (in)formal collusion, preferential tax treatment, trade and investment pacts, the political influence on e.g. labor and financial regulation, trademarks and product differentiation, and manipulative advertising. Together, they massively decrease productive efficiency e.g. by phony product differentiation (1919b, Chapter 3). Veblen (1921) only ironically proposed to counteract sabotage with the formulation of the opposite goal, i.e. a maximization of output. This would contradict his criticism of wasteful mass consumption and his ecological concerns. Instead, he forces the readers to ask themselves where the limits of production and growth should be not least in view of decreasing average costs and not early increasing marginal costs and natural limits of firm sizes. Lamoreaux (1985) highlighted in her historical analysis that the policy of maximum production had fatal consequences for firms at the turn of the 20th century. Veblen’s diagnosis of “excessive productivity”, thanks to technological advance, captures a feature of overproduction in many mass production branches like the automobile industry of today. Against the argument, that competition will force companies to reduce prices to the level of the production costs or to increase production given decreasing average costs, Veblen argues that the continued progress of the industrial arts has become a continued menace to the equilibrium of business, has forever threatened to lower the cost per unit and to increase the volume of output beyond the danger point, - the point written into the corporation securities in the shape of fixed charges on funds borrowed for operation under industrial conditions that have progressively grown obsolete. (1923, 97) Veblen’s point is that increased credit, leverage, and financing via stocks and the obligation to pay dividends necessitate “high” prices to be able to pay back credit, interest, and dividends in addition to depreciation and normal profits. The more developed the credit system, the higher must be the prices, depending on the share of credit per unit of output. Such loans cannot, at least not directly, swell the aggregate industrial equipment or enhance the aggregate productivity of industry; for the items which here serve as collateral are already previously in use in industry to the extent to which they can be used … To a very considerable extent the funds involved in these loans, therefore, have only a pecuniary (business) existence, not a (material) one. (1904, 54) Capital so far has a dual character as large-scale physical productive capacity and as financial predatory power. Veblen accentuates that the expansion of the credit system does not necessarily enable and facilitate productive investments, but diverts production, distracts from real investments, and drives up prices. Veblen’s contributions (especially 1904) convey a critique of a progressively decoupled financial sphere from the real economy, including a colorful description of financial jugglers and speculation. His criticism has a very familiar ring after the financial crisis of 2007ff. (Wray 2012 and Peukert 2013, Chapters 2.2 and 2.3). 36

Thorstein Bunde Veblen

The credit system is always in an unstable equilibrium: A boom starts with increased earnings, and extension of credit since the nominal earnings exceed the loan liabilities, which are fixed in price. There results a growing discrepancy between “real earning capacity” and the capitalization of the collateral for the loan. When the divergence between the value of the business capital and its actual industrial earning capacity becomes “exceptionally wide” and “the overrating is presently recognized by the creditor and a settlement ensues,” business capital is reduced to the actual earning capacity, that is, the price of the firm’s stock falls and the loan expansion stops. (Phillips 1989, 695; the quotes refer to Veblen 1904) Veblen shared and deepened a widespread rejection of economic and political concentration, trusts, and impersonal joint stock companies of his time (with reference to Veblen see e.g. Supreme Court Judge Louis Brandeis’ criticism 1914). As a counterpoint to commencing mass producer and consumer society, he appreciated the Jeffersonian vision of the self-reliant common man. Veblen’s positive utopia is hidden in Imperial Germany (1915) and The Higher Learning in America (1918). His ideal – there at least – was “pagan anarchy”, with more autonomous and to a preferably high degree self-sufficient citizens, including agricultural producers and a non-hierarchical direct democracy.

2.5

Veblen as a precursor of evolutionary economics and his current significance

Veblen broached the main issues of modern evolutionary economics: He discussed entrepreneurship, technical change, the coordination of human actions in groups, and the change of needs and wants. He criticized the structural minimalism of equilibrium economics and methodological individualism. Instead, he analyzed processes, adaptive structures, and cumulative causation (Berger (Ed.) 2009). He pleaded for pluralism in method, the inclusion of neighboring sciences like sociology, and concrete and ostensive case studies. Epistemologically, he highlighted the limits of knowledge in principle and in the context of complex societal change and on the level of group selection. This included all symbolic interpretations and the sciences. He highlighted the role of implicit knowledge and modular-evolutionary reason and the influence of socialization (imprinting) and thereby overcame a rationalist theory of cognition. Besides (anyway embedded) market exchanges he considered the role of norms, routines, formal and informal rules, law, the striving for market power (including patents and intellectual property rights), as well as political authority and coercion as mechanisms for replication. He defined institutions as cognitive frames of what evolutionary economists today call “network configurations”. With the Veblen-, the bandwagon-, and the snobeffect, he came close to the insight of complex systems behavior (chaos theory). Reflexes and instincts were phylogenetic results of natural selection vis-à-vis the material living conditions, co-evolving with culture. Development is for him a never-ending process in real time on an irreversible time path (non-ergodicity, historicity). He distinguished production regimes and systems and specified long-wave technological shifts and bifurcations. He discussed the technostructure as a whole and as a collective enterprise with positive external effects on production and negative real and financial external effects are concerned. Constant rivalry to appropriate the returns with a pecuniary-predatory attitude prevails in present-day economies, discussed in modern evolutionary economics under the 37

Helge Peukert

headings of principal-agent conflicts, closure of networks, erection of champions, application of strategies especially according to non-cooperative game theory, the hawk-dove game strategy of the bourgeois, etc. He attributes due weight to creativity (idle curiosity), hazards, and surprises (Peirce), counterbalanced by institutional and mental inertia and redundancies (lock-ins; on cultural lags see the empirical Veblenian analysis of Marquez 2019). Veblen’s strong emphasis on conspicuous consumption and emulation is also a theme in modern evolutionary economics in the shape of the handicap principle, status preferences, positional competition, difference games, bowling alone, anomie, and e.g. winner-take-all concepts (for these basic concepts of evolutionary economics see Herrmann-Pillath 2002 and Nelson et al. 2018). But Veblen’s dealing with the main cultural and economic evolutionary processes and concepts is particular, as his oeuvre is also an attack on dominating institutional and mental preconceptions. He questions, (1) that there is a fundamental harmony of interests between labor and capital, (2) that the pursuit of individual self-interest will lead to the overall social benefits through the workings of the Invisible Hand, (3) that competition is the most efficient and equitable way to reward merit or competence, (4) that the state is essentially a balance wheel, keeping many and varied interest groups in a state of equilibrium with each other, and (5) that the main conflict is not between man and man but between man and nature. (Tilman 1985, 881–882) Veblen’s approach of combining technostructure, scientification, capital as common knowledge, and tight corporate-government alliances was further developed by Galbraith (1967) and profoundly applied theoretically and empirically to the present age by Nitzan and Bichler (2009) whose starting point is explicitly Veblenian (on entrepreneurial sabotage see their Chapter 12). Bichler and Nitzan substantiate that when a lot is invested, the valuation of companies on the capital market tends to fall. On the other hand, if little is invested, the company’s value increases. Greenwald et al. (2020), for example, found that up to 1988, 92% of company value increases could be statistically explained by rising productive value added, but only a quarter in the last 30 years. Rather, since then more than half of the increase in value was caused by the redistribution of economic rents. Microsoft and Apple, for example, are excellent examples of rent appropriation in surveillance capitalism. In 2005, Microsoft’s buildings and facilities, worth US$2.3 billion on the balance sheet had a market value of $283 billion. Including debts, little more than half a percent is due to the value of the means of production. Industries with huge patent packages resemble minefields for small and medium-sized companies. Only the largest companies can afford to operate in these industries. The profit margins are then correspondingly high. Mobile telephony is one example. Apple’s market value in 2020 has risen from one to two trillion euros within months. With a rent appropriation approach, research in Veblen’s critical institutional tradition could contribute to the multifaceted evolutionary study of firms, industrial dynamics, network effects, and industrial evolution (see the examples in Pyka and Dopfer 2019, Volumes 2, 2.3, and 2.4). Veblen saw no natural limits to mergers and concentration in the face of increasing returns in mass production so that antitrust policy is warranted. John Maurice Clark’s approach to workable competition, based on Veblen, had an effect right up to the competition policy model of West Germany. 38

Thorstein Bunde Veblen

As evidence of Veblen’s influence on the social sciences, reference can be made to Bourdieu’s Distinction (2010/1979) on the subtle differences, who analyzed the respective dominant social habits and cultural capital as an independent dimension of the exercise of power alongside other forms of capital. From a Veblenian perspective of social comparison, Richard Frank (e.g. 1999) empirically examined the consequences of increasing inequality, which leads to increasing frustration of the multitude. Frank vividly describes the luxury of today’s American upper class and suggests a workable, progressive consumption tax. From an economic psychological point of view, happiness research (Binswanger 2008), and parts of behavioral economics (Kahneman 2011, especially Chapter 37) confirm Veblen’s key assumptions about the treadmills of happiness. His characterization of the predatory state makes the conspicuous behavior and success (not only) of Donald Trump understandable (Plotkin 2018). In the face of the global ecological crisis as today’s main challenge for human evolution, the pursuit of luxury and the consumption of status are at disposition. Claims for a sufficient economy that tremendously reduces total resource consumption are being increasingly raised (Kern 2019). Veblen already detailed in the chapter “The timber lands and the oil fields” (1923, 186–201) the “routine of waste and inefficiency” (1923, 200). He especially mentions deforestation and the careless overuse of natural resources. Veblen’s criticism of irrational status consumption, unfair distribution, and the tenacity of outdated institutions and habits of thought and his decentralized-sufficiency ideal with elements of macroeconomic and technological control (Kapp 2016) lives on in drafts of the post-growth economy (Mitchell (Ed.) 2007).

References Atkins, Willard. “Comment”. In: American Economic Review (Supplement), 22, 1932, 111–112. Berle, Adolf & Means, Gardiner. The Modern Corporation and Private Property, New York 1932. Berger, Sebastian. (Ed.). The Foundations of Non-Equilibrium Economics: The Principle of Circular and Cumulative Causation, Abingdon 2009. Binswanger, Mathias. Die Tretmühlen des Glücks, 3.Aufl., Freiburg 2008. Bourdieu, Pierre. Distinction, London 2010 (1979). Brandeis, Louis. Other People’s Money, New York 1914. Camic, Charles/Hodgson, Geoffrey. (Eds.). Essential Writings of Thorstein Veblen, London 2011. Dopfer, Kurt. Thorstein Veblens Beitrag zur ökonomischen Theorie“. In: Grüske, Karl-Dieter (Ed.), Vademecum zu einem Klassiker des institutionellen Denkens, Düsseldorf 2000, 89–144. Frank, Robert. Luxury Fever, Princeton 1999. Galbraith, John Kenneth. The New Industrial State, Boston 1967. Greenwald, Daniel et al. How the Wealth was Won. NBER Working Paper, No. 25769, 2020 ( https:// www.nber.org/papers/w25769). Herrmann-Pillath, Carsten. Grundriß der Evolutionsökonomik, Weinheim 2002. Hodgson, Geoffrey. Economics and Evolution, Ann Arbor 1993. Hudson, Michael. (Ed.). Absentee Ownership and Its Discontens, Dresden 2016. Jorgensen, Elizabeth & Jorgensen, Henry. Thorstein Veblen: Victorian Firebrand, New York 1999. Kahneman, Daniel. Thinking, Fast and Slow, New York 2011. Kapp, William. The Heterodox Theory of Social Cost, ed. by Sebastian Berger, Nottingham 2016. Kern, Bruno. Das Märchen vom grünen Wachstum, Zürich 2019. Lamoreaux, Naomi. The Great Merger Movement in American Business, 1895–1904, Cambridge 1985. Marquez, B. From the Peaceable to the Barbaric: Thorstein Veblen and the Charro Cowboy, New York 2019. Mitchell, Ross (Ed.). Thorstein Veblen’s Contribution to Environmental Sociology, Lewiston (NY) 2007.

39

Helge Peukert Nelson, Richard et al. Evolutionary Economics: An Overview, New York 2018. Nitzan, Jonathan & Bichler, Shimshon. Capital as Power, London 2009. Peukert, Helge. Das Handlungsparadigma in der Nationalökonomie, Marburg 1998. Peukert, Helge. “On the origins of modern evolutionary economics: The Veblen legend after 100 years”. In: Journal of Economic Issues, 35, 2001, 543–555. Peukert, Helge. Die große Finanzmarkt- und Staatsschuldenkrise, 5., rev.ed., Marburg 2013. Phillips, Ronnie. “The Minsky-Simons-Veblen connection: Comment”. In: Journal of Economic Issues, 23, 1989, 889–891. Plotkin, Sidney. Veblen’s America: The Conspicuous Case of Donald J. Trump, London 2018. Pyka, Andreas & Dopfer, Kurt (Eds.). Evolutionary Economics. 4 Volumes. London 2019. Tilman, Rick. “The Utopian vision of Edward Bellamy and Thorstein Veblen”. In: Journal of Economic Issues, 20, 1985, 879–898. Veblen, Thorstein. “Why is economics not an evolutionary science?”. In: Quarterly Journal of Economics 12, 1898, 373–397. Veblen, Thorstein. The Theory of the Leisure Class. New York 2007 (1899). Veblen, Thorstein. The Theory of Business Enterprise, New York 1904. Veblen, Thorstein. The Instinct of Workmanship and the Industrial Arts, New York 1914. Veblen, Thorstein. Imperial Germany and the Industrial Revolution, Ann Arbor 1915. Veblen, Thorstein. The Higher Learning in America, New York 1918. Veblen, Thorstein. The Place of Science in Modern Civilization, New York 1919a. Veblen, Thorstein. The Vested Interests and the State of the Industrial Arts, New York 1919b. Veblen, Thorstein. The Engineers and the Price System, New York 1921. Veblen, Thorstein. Absentee Ownership, New York 1923. Veblen, Thorstein. The Collected Works of Thorstein Veblen, 10 Volumes, London 1994. Wood, John Cunningham (Ed.). Thorstein Veblen: Critical Assessments, 3 Volumes, London 1993. Wray, Randal. “The great crash of 2007 viewed through the perspective of Veblen’s Theory of Business Enterprise, Keynes’s Monetary Theory of production and Minsky’s financial instability hypothesis”. In: Reinert, Eric & Viano, Francesca (Eds.), Thorstein Veblen, London 2012, 303–316.

40

3 THE FOUNDATIONAL EVOLUTIONARY TRAVERSE OF RICHARD R. NELSON AND SIDNEY G. WINTER Isabel Almudi and Francisco Fatas-Villafranca 3.1 The context The principle that marginal subjective valuations and their interpersonal confrontation in exchanges are fundamental aspects of market order and relative price formation was almost simultaneously stated by William Stanley Jevons, Carl Menger, and Leon Walras in the early 1870s. This principle is now regarded as one of the pillars of modern economics. During the 19th century, the objective value theory of the Classics had been questioned. The scheme pioneered by Adam Smith and David Ricardo, which reached a systematic presentation in John Stuart Mill’s Principles of Political Economy, had become the object of critical attacks from different angles: from the German Historical School and the continental schools of law and philosophy; from alternative views that existed in France and Italy (Cournot, Dupuit); but also in England, where Alfred Marshall knew well the work of von Thünen and had begun thinking of an improved and combined supply-demand scheme. These 19th-century debates led to the configuration of the marginalist revolution with their subjective value theory. In a few decades, the role of objective costs from the supply side was also well integrated in the theory, and the impact of the modern approach was permeating, in its distinct variants, wide realms of political economy. We find it in the work of Wicksell and the Austrian economists Wieser and Böhm-Bawerk, and it engenders the neoclassical paradigm from Pareto and Edgeworth to the market theory of supply-demand equilibrium in Pigou and Marshall. Of course, there were remarkable exceptions: Karl Marx worked along classical paths to establish his Hegelian-but-materialist revolutionary view (enormously influential during the 20th century after the Russian revolution). And from a very different viewpoint, the Old American Institutionalist School with Thorstein Veblen as the leading figure, developed its theory of social groups, focusing on concrete habits and lifestyles and studying the origins of economic institutions. It was precisely Veblen, the first who wondered whether economics, perhaps, should be considered an evolutionary science (Veblen, 1898). This could be a brief description of the theoretical landscape at the beginning of the 20th century, from which the Neoclassical and Austrian schools of economics (with their remarkable differences) ended up prevailing as the mainstream approaches, at least during the first two decades of the century. Notwithstanding the importance of social categories DOI: 10.4324/9780429398971-5

41

Isabel Almudi and Francisco Fatas-Villafranca

and market processes recognized by economists at the time, the inquiry in both schools was primarily posed in terms of individual agents organizing their allocative action around subjective perceptions and budget constraints, and interacting with profit-led firms constrained by market prices and the state of technique (Robbins, 1932). On the Austrian side, Mises and Hayek were extending their frame to deal with the monetary causation of cycles; in the neoclassical realm, the rigorous and rich analysis of Marshall and Pareto was quickly integrated by John R. Hicks (1939). Favorable to the use of mathematics in economic analysis, the neoclassicals fostered the use of linear algebra and differential calculus in the synthetic comparative static methodology conveyed by Samuelson (1947) (relying on the implicit function theorem, classic constrained optimization techniques, and the correspondence principle for dynamical adjustments). As we can see in Allen (1960) this central core of microeconomic inquiry was culminated by WWII. Drawing on these mainstream schools, but moving in different doses away from them, we observe in the 1930s six strong theoretical breakthroughs that were to shape 20th-century economics. These breakthroughs conformed to the training and early context of Nelson and Winter. The first novelty was that a small research program which began as a technical problem (trying to delimit the conditions for a Walras-Cassel equilibrium to exist) became (through the framing of Wald, the technical intuition of von Neumann, and the support of Divisia and Allais) the theoretical core of neo-Walrasian general equilibrium theory (GET). This research gained its central place after the war with Arrow and Debreu (1954) and Debreu (1959), all the way to Dierker (1974). These authors solved correctly the technical problems left by Walras and Pareto and broke the previous compromise of neoclassical economics with differential calculus. More precisely, by extending the implementation of convexity theory, separation theorems, and topological methods in neo-Walrasian settings, Arrow, Debreu, and McKenzie found conditions to explain: (i) how prices for present and future markets could be instantaneously determined, from the coordination of demand-supply (consumption-resources) plans (devised by axiomatically characterized maximizing consumers in private ownership economies), and supply-demand (output-input) plans (devised by profit maximizing firms owned by consumers); and (ii) how are these market equilibrium prices, related to Pareto-optimal states of neo-Walrasian economies. The Arrow-DebreuMcKenzie model was path-breaking in neoclassical theory, it was part of Nelson and Winter’s early career, and it turned out to be – through Arrow and Hahn (1971) – a key reference for economists in the second half of the 20th century. The second innovation was mostly technical. During the thirties, we can observe the prefiguration of econometrics and quantitative statistical methods as the privileged tool for contrasting economic theories. The work of Kuznets (1929), the econometric views of Tinbergen (1939), Frisch (1933), Koopmans (1936), Haavelmo (1943), and, later on, the advances within the Cowles commission oriented by Koopmans and colleagues in Koopmans (1950), all established the classic econometric program which necessarily influenced the training of our protagonists. A reference to understand the econometric program in the fifties and sixties, as well as how econometrics have benefited in recent decades from the advances in electronic devices, may be Theil (1978). In this text, we can see how inferential statistics and the testing of parametric hypothesis, multiple regression analysis and maximum likelihood, confidence intervals, estimation, validation, prediction, and control were all implemented to quantify and test the economic theories of the time. It is also remarkable that an alternative quantitative line of research at the Cowles foundation was developed in different directions from the models of Leontief. Thus, Koopmans, 42

The foundational evolutionary traverse of Nelson and Winter

Dantzig, Simon, Samuelson, and Solow (Koopmans, 1951) introduced linear programming and activity analysis in economics, influencing the early ideas of Nelson and Winter. The third theoretical breakthrough of the interwar period was the work of John M. Keynes (1936), which was very much induced by the 1930s economic crisis and considered the source of many analytical developments that followed. Although, perhaps, the truly path-breaking pieces in Keynes’ General Theory were the role of radical uncertainty in explaining the volatility of expectations and investment, the trembling nature of the demand for money, and the fragility of effective demand in maintaining aggregate equilibria with full employment (Keynes, 1937; Alchian, 1955; Shackle, 1979), it must be recognized that the most influential immediate interpretations of Keynes were not so far from the mainstream. We refer to the works by Alvin Hansen and, mostly, to the article by John R Hicks (1937) – the IS-LM model, later extended with the mathematical tools provided by Samuelson (1947). This approach led to the Keynesian-Neoclassical synthesis of Modigliani, Tobin (1955), Blinder and Solow (1973) and the large econometrically estimated models for policy in Klein and Goldberger (1955). Even the early Milton Friedman (1957) was involved in these extensions. This interpretation of Keynes was of paramount importance in establishing macroeconomics as a field, and in searching for the micro-foundations of consistent macro-models that oriented policy (Tobin, 1955, 1969). They considered Keynesian pathologies as problematic temporary states in the short run that, once the prescribed stabilization policies were implemented, could be amended to restore the general equilibrium neoclassical predictions in the long run (Solow, 1956; Tobin, 1965; Phelps, 1970). Of course, there was an alternative reception of Keynes’s contribution by some direct students and collaborators of Keynes in Cambridge. These were the so-called post-Keynesians. They perceived in Keynes a much stronger critique of the functioning of market economies, as we can see in the works of Joan Robinson, Kalecki, Harrod, Kaldor, or Saraffa. Although we may note that their works were, perhaps, as tributary to Keynes as they were to Marx or Ricardo, their interpretations pointed out essential phenomena, such as dynamic increasing returns, structural change, and imperfect competition; these works led to the growth models of Pasinetti (1981) and have become very influential in evolutionary economics (Shiozawa et al., 2019). The fourth notable development in the interwar period has to do with the renewal and relocation of the leading Austrian thinkers. Hayek moved to LSE and much later to the University of Chicago. He developed, in fruitful essays, the notions of dispersed, tacit, and practical knowledge, as well as the idea of spontaneous orders in capitalism driven by entrepreneurial discovery processes, price information, and market interactions. In Hayek (1937, 1945, 1967), the scattered and non-fully articulated nature of knowledge in individuals, markets, and organizations and the role of prices as coordinating signals are coprefigured in relation (and friendship) with Michael Polanyi (1962, 1967) and Karl Popper (1945). Meanwhile, Mises (1949) finished in New York his monumental Human Action, in which he defines in depth the position of the Austrian school of economics (in critical controversy with Lange and others) regarding the Socialist calculation debate. Drawing on his praxeology and extending previous Austrian insights, Ludwig von Mises explores the multidimensional set of wants and capabilities co-existing in modern societies and the enormous informational and computational requirements that a society must face to cope with the income generation and distribution questions of what, how, and to whom produce. In relation to this, it is also during this period that Joseph A Schumpeter moved to Harvard. In the USA, Schumpeter completed his initial work on entrepreneurial innovations and their disruptive effects (Schumpeter, 1934), and came up with his results on long-cycles, and 43

Isabel Almudi and Francisco Fatas-Villafranca

on the routinized character of corporate R&D. It is interesting, on the one hand, how Schumpeter vindicates in this period the figure of Karl Marx; but, on the other hand, he demolishes key Marxist conceptions (Schumpeter, 1942). In his Capitalism, Socialism and Democracy, Schumpeter argues that, in analyzing capitalism, we are dealing with an evolutionary process. The fifth innovation of the period is a new stream of institutional thought that, focusing on organizational, legal, social, and non-market aspects and showing a remarkable interest in economic history and the study of cases, was to shape with time the New Institutional Economics (Coase, 1937; Williamson, 1975, 1985; North, 1991, 2003). While maintaining a certain tension with the atomistic market equilibrium approach that we have found in neoclassical economics, but also delimiting clear differentiations with the (exclusively) marketprocessual view of the Austrians, this school obtained interesting results regarding the following themes: the detailed study of multifaceted organizations and the problems of uncertainty in modern societies (Knight, 1921, 1951; Alchian, 1950); the nature of firms and its peculiarities, boundaries, and motivations in particular sectors (Coase, 1937); the problems of transaction costs, social cost vs private costs, and market miss-allocations (Coase, 1960); property rights, externalities, and information problems (Alchian, 1950; Demsetz, 1967; Alchian and Demsetz 1972). This approach will be influential in Nelson and Winter, and more so when combined with the Carnegie School. The final breakthrough from the interwar period, which conditioned the context of Nelson and Winter, came from certain parallel advances that happened in mathematics during the 1930s, 1940s, and 1950s. We refer to some developments in the mathematics of evolution and in Game Theory. These advances became, to a certain extent, partially synthesized in the 1980s with the work of John Maynard-Smith (1982) – Evolutionary Game Theory. We can begin by noting that one of the most remarkable scientific advances during the first quarter of the 20th century, was the integration and unification of Mendelian genetics and Darwinian principles in mathematical biology. Regarding the influence on our protagonists, the seminal work of Ronald Fisher (1930) conveyed formulations of concepts such as selection, replication, and stochastic mutation that played a role in evolutionary economics. Posterior studies on covariance selection mathematics by Price (1972), and in replicator dynamics [synthesized in Schuster and Sigmund (1983) or in Hofbauer and Sigmund (1998)], were also influential in Nelson and Winter and their school. From a different angle, but in a partially convergent direction, John von Neumann and Oskar Morgenstern (1944) posed their theory of strategic behavior in the book Theory of Games and Economic Behavior. They provided just the beginning of the theory. Probably, it was through the work of John Nash in the 1950s, who formulated his notion of equilibrium and applied topological tools and fixed point theorems to prove that every finite game has an equilibrium, that game theory became consolidated. Although John Nash suggested a possible mass-population evolutionary interpretation of his equilibrium concept, it was not until the work of Maynard-Smith in the 1970s and 1980s that the attention of game theorists turned away from refining notions of rationality and equilibria to the study of learning. We want to anticipate that these mathematical instruments were useful for our protagonists. We find them among the bundle of technical resources that Nelson and Winter used when they began modeling populations of heterogeneous agents displaying differential growth rates, with the agents being rule-driven, random innovations appearing, and the levels of market competitiveness being seen as analogous to fitness levels. For the time being, we leave here the context and move to the study of the initial questions and problems that our protagonists faced. 44

The foundational evolutionary traverse of Nelson and Winter

3.2 Economic order and firm theory in the early Sidney Winter The article by Alchian (1950) condenses three important preliminary insights which, taken together, will mark the early efforts of Sidney Winter. In this article, Armen Alchian rises, firstly, the question of how economic theory can deal with the problem of market order and price formation in continuously changing and radically uncertain environments; secondly, he suggests that a fruitful way to tackle this problem, without dispensing entirely with mainstream economics, may be studying firm adaptation processes and their outcomes in the long run; and, thirdly, Alchian proposes that the methodology to face these challenges relies on the combination of economic theory with the theory of stochastic processes (Feller, 1957). The third (methodological) suggestion was taken seriously by mainstream economics, but the first and second points were largely ignored (Stokey et al., 1989). On the contrary, the young Sidney Winter took the three challenges altogether and began addressing the problems of market functioning, price formation, and planning in multisector settings and how firm adaptation efforts may cope with uncertain dynamic environments. In order to understand how our author was gradually facing and combining his thoughts on the three Alchian challenges, we must primarily figure out (through the early texts) which were the skills, direct research contacts, and preliminary explorations of Winter. Regarding the overall context presented in the previous section, we see in the texts how Winter becomes interested in, at least, four of those lines of advance. The first line consists of a profound knowledge and (critical) interest in the nascent literature on neo-Walrasian GET. This is clear in Winter (1969), but also in mentions by other authors – such as Hal Varian (1980) – who, in the preface for his advanced microeconomics textbook, recognizes that the chapters on General Equilibrium Theory, Welfare Economics, and the Theory of the Firm all draw on the lecture notes by Sid Winter and Daniel McFadden at the University of Michigan. Both in Michigan, and later at Yale University, Winter shows an advanced knowledge of the neo-Walrasian approaches and techniques. Apart from his comments on the fundamental theorems of welfare economics, it is remarkable his familiarity with the literature on preference aggregation and axiomatic social choice, as established by Arrow (1951) and later Sen (1970). This is very important because he will end up (much later) thinking against this tradition, that he perfectly knows. Secondly, and this is surely due to the proximity of Winter to Koopmans, Simon, the Cowles Foundation (already established at Yale in those years), and RAND, Winter was highly involved in the frontier research of the time on representing technology in multisector Activity Analysis models. The influence of Koopmans, Dantzig, advances in convexity theory, linear programming, multisector linear technology settings, and the Kuhn-Tucker results in the fifties (partly influenced by Game Theory), is enormous in the young Winter, as it can be seen in Winter (1965, 1967; and later in 1981, 1982). This is essential at least for two reasons: first, because it will influence, at least partially, the posterior characterization of technology and technological advance in evolutionary models with Richard Nelson. And also, second and very remarkable, because of the contributions of Winter to: i) the literature on turnpike paths in von Neumann (1945) and Radner (1961) models; ii) to the study of linear economies with many activities á la David Gale (1960; see also 1955, 1956); and iii) to the assimilation of linear programming as a development of Leontief analysis – as in Koopmans (1951), or in Dorfman et al. (1958), all this kept Sid Winter significantly away from the Macroeconomic Keynesian aggregate models of the time. This is essential, because it reveals, in the young Winter, a conviction with regard to the need of encompassing

45

Isabel Almudi and Francisco Fatas-Villafranca

multiple interconnected activities and multisector settings to explore economic order in large economies. This will become later (with Nelson) a strong interest in structural change; and it shows a perception of the fact that the mainstream translation of microeconomics into macroeconomics, through typological thinking in permanent (even moving) equilibrium, is a closed way. Thirdly, Sidney Winter presents an increasing passion for methodological discussions (Alchian, 1950; Friedman, 1953; Koopmans, 1957) that will eventually lead him to a deep interest in very foundational questions such as what is economics as a science and how to achieve empirically significant advances in economic theory. This will become very clear in later works (Nelson and Winter, 1982; Nelson and Winter, 2002). Finally, fourth, from the study of the three Alchian challenges and the reading of Friedman on the methodology of positive economics, Winter (1964) comes up with a critical view of the characterization of firms as profit-maximizers on well-defined technology-choice sets. He progresses gradually towards the assimilation of the behavioral theory of the firm (Simon, 1947; Cyert and March, 1963). These topics are in line with the institutionalist concerns in the thirties, and connect directly with the contributions of the Carnegie School to organization theory, and the notion of bounded-rationality by Simon (1957, 1983). These interests are clear in Winter (1964) but reach scale in the 1970s. In Winter (1964), the author faces the Alchian challenges and the Friedman defensive argument by wondering what it means to consider firm theory as a theory on how firms adapt their behavior to market changes. By taking seriously the Friedman conjecture that: i) firms react to market conditions; ii) the survival of the fittest and the exit of unviable firms clean over the initial set of firms; and iii) therefore the fittest and prevailing firms in the long run should be those behaving as if they were profit-maximizers, Sid Winter realizes that the problem is not that simple. To begin with, in theorizing about the firm we must consider, at least, three key processes: 1 A cognitive process through which firms acquire information dependent on the state of the world but conditioned on their specific organizational form. 2 A decision process, which is dependent on the information acquired but also on the internal conditions of the organization. Processes 1) and 2) can be formally captured in a typical Marschak and Radner (1972) formulation. 3 Processes of entry-exit in which the appearance-disappearance of specific identifiable firms is paired with their concrete organizational traits conditioning 1) and 2). This is very important because the Friedman argument seems to privilege the role of market exit, but it does not consider well the entry of new organizational variants. Moreover, Friedman did not distinguish in his “natural selection” defense of profit maximization between actions and organizational forms. It is remarkable that in his 1964 essay, Winter already combines behavioral and organizational ideas (as those presented above) with proto-evolutionary formulations, and even he provides us with a preliminary notion of firm routine. These ideas, and once the collaboration with Richard Nelson has slightly begun, appear developed in the magisterial paper by Winter (1971) in the Quarterly Journal of Economics. In this article, Sidney Winter’s combination of behavioral, evolutionary, technological, and mathematical new ideas appears perfectly interlinked. Drawing on the behavioral critique that orthodox firm theory neglects the real decision mechanisms in firms, Winter argues that to understand the behavior of 46

The foundational evolutionary traverse of Nelson and Winter

adaptive organizations in rapidly changing markets, we must study its internal structure, decision rules, and the concrete mechanisms of adaptation. In the fifth section of the chapter, employing the theory of finite Markov chains, the author proposes a one-industry evolutionary model in which heterogeneous boundedly rational (satisficing) firms operating with rules, displaying different growth rates (transition probabilities from t to t+1 towards alternative states including growth, entry, exit), and facing a standard demand side of the market, engender a competitive process formalized as a Markov process on the set of industry states. Then a strong formal result is proved on the possibility of coming up with a realistic firm and industry dynamics theory capable of understanding firm behavior and industrial change in non-equilibrium conditions. The theory of stochastic processes used for firm transition probabilities, the incorporation of an original and tractable evolutionary model embodying everything learned on the representation of techniques and organizations, the consideration of firm entry-exit, and a peculiar characterization of demand, all led Sidney Winter to establish sufficient conditions under which, the long-run equilibrium of the modeled industry, would be identical to a (conveniently tuned) typical outcome of partial equilibrium neoclassical models (although no firm in Winter model may be maximizing profits). Moreover, by removing one or several of the sufficient conditions, we observe industrial dynamic paths in the model leading to non-neoclassical configurations. It is a theorem that proves formally the need to differentiate between the assumption that firms maximize profits, and accepting the possibility of potential particular outcomes in which firms may end up being (or not) profit-maximizers in industry models. Besides all this, we may note that the demand side of Winter (1971) industry analysis, as characterized by his seventh assumption, already anticipates certain shortcomings regarding demand that weaken the first epoch of modern evolutionary economics. Basically, demand plays here the role of closing the system, as will be the case in Winter (1984), with no evolutionary insights in this respect. This is a typical shortcoming of early neo-Schumpeterian evolutionary economics (and of Schumpeter himself) that is being improved in recent decades (Witt, 2001; Valente, 2012; Chai and Baum, 2019). It is, nevertheless, interesting to note here that an insightful alternative manner of dealing with demand from an evolutionary perspective had been tried earlier by Phelps and Winter (1970), where Winter presents a customer-flow revision protocol in an evolutionary approach to demand, but – in this case – combined with a neoclassical supply side (a dynamic optimization problem for price-setting firms). Recent advances in evolutionary economics, such as Almudi et al. (2013) and Almudi et al. (2020) in Industrial and Corporate Change, have combined the evolutionary supply side in Winter (1971) with the evolutionary demand side in Phelps and Winter (1970). The combination produces surprising results on coevolution in multisector settings and industrial policy. In order to close the early research by Sid Winter, we note that in Winter (1982, 1986) the author explores the notion of rationality, the nature of the firm, organizational capabilities, and possible rigidities arising when organizations display adaptive behavior, thus anticipating later collaborations with the neo-Institutional views, as in Williamson and Winter (1991). Likewise, in relation to the problems of information acquisition and large-scale social organization (that concerned our protagonist at least since his studies on turnpike theorems and social planning), Winter warns us against the price mechanism as being considered the unique and even the most effective information transmission activator in complex economies. This theme also appears in Arrow (1974), and in Winter takes the form of exploring the low elasticity of substitution in firms, when facing changes in relative input prices (Winter, 1981). During the eighties, we see in the author highly elaborated arguments 47

Isabel Almudi and Francisco Fatas-Villafranca

on the nature of knowledge, revealing how Winter was becoming influenced by his collaboration with Nelson.

3.3

The problem of economic change in the early Richard Nelson

Richard Nelson always explains that the program in economics at Yale University and elsewhere in the early 1950s (his training years), was much richer in contents related to economic history and economic thought than it is today. This familiarity with historical studies, with the history of thought (Smith, Ricardo, Marshall, Schumpeter), together with the early involvement of Nelson in field studies related to technological progress (one year of engineering at MIT, and collaborating with technologists at RAND), conditioned his interest in economic change. Of course, the early Nelson was also influenced by the prevailing theoretical approaches that we have seen in Section 3.1. It is clear in Nelson (1956, 1964) that he was thinking (at least in part) from the Keynesian-Neoclassical synthesis of James Tobin and colleagues, and he was doing growth accounting with the new date series, drawing on aggregate neoclassical production functions. But, even in these early works, Nelson dealt with inter-disciplinary frontier issues that, with time, led him far from the mainstream coordinates. For example, in his study on low-level equilibrium traps for low-income economies, Nelson (1956) already detects the role of the socio-cultural environment as a potential engine of economic growth (under certain conditions), even if capital per capita was not growing at the low-level specific trajectories. This analysis anticipates that Nelson was going to perceive the processes of economic growth and development, as being much more complex phenomena than the accumulation trajectories obtained in Solow (1956, 1957), in the optimal-growth Ramsey-Cass-Koopmans models, or even in the overlapping generations models being proposed at that time (Cass, 1965; Samuelson, 1958; Diamond, 1965). We claim that Nelson’s conception is more complex because it considers a much wider array of non-market phenomena, notwithstanding the relative advances in Ramsey-CassKoopmans (they consider intertemporal preferences combined with investment and the productive structure in order to characterize optimal consumption, capital, and income growth paths); or the advances from Samuelson’s overlapping generations model with money, which later allowed Peter Diamond (1965) and colleagues to study growth with a portfolio of assets. These aspects had already been studied, but from a different angle, in the Keynesian Tobin-growth models with money. On the other side, it is remarkable in Nelson that, even when he uses terse standard models or growth accounting techniques (Nelson, 1964, 1968, 1981, 1982 much later Nelson and Pack, 1999), he always operates clearly influenced (in his tendency to encompass many pragmatic details of corporate activity and policy-making) by a small group of young economists and technologists of his generation (Zvi Griliches, Edwin Mansfield) and by his partners at the US Council of Economic Advisors (James Tobin, Arthur Okun, Kenneth Arrow) who, because of the nature of their tasks at that moment, focused on practical implementation contexts. They considered the specificities of distinct real sectors (agriculture, aeronautics), concrete historical episodes, and tried to figure out the impact of their ideas on social organization. Of course, this global and mixed Nelsonian conception of economic practice is, somehow, at odds with the prevalence of highly stylized models in the mainstream, and as early as in his paper on technology diffusion (Nelson, 1968) or in his joint study with Winter on weather forecasting and information (Nelson and Winter, 1964), 48

The foundational evolutionary traverse of Nelson and Winter

he realized that removing key assumptions of mainstream models could be essential for studying economic change. Regarding other streams of thought framing early Nelson’s ideas, let us briefly bring out two additional influences that followed from the interwar breakthroughs that we mentioned in Section 3.1. Firstly, a very important circumstance was the contact of our protagonist with Burton H. Klein, a former student of Schumpeter, who in Klein (1977) presents a fresh view on Schumpeterian competition, the generation of knowledge, and the transformation of national capabilities, that had permeated young people (like Nelson and colleagues) at RAND in the previous decade. The influence of Schumpeter, and of Schumpeterian ideas were to become a permanent mark in Nelson thought, and operated as co-activators of the evolutionary theory of capabilities that Nelson and Winter developed in the 1970s and 1980s. Secondly, another influence following from the interwar breakthroughs, that shaped Nelson thinking, was that of the organizational theorists and institutional economists (all the way from Simon, 1947 to Alchian and Demsetz, 1972). This stream enriched the Nelsonian analysis of technological and economic change with elements related to legal systems, property rights, the nature of organizations, the notion of rationality, and aspects of social action in general. The peculiar style of the young Richard Nelson and his familiarity with real cases and the history of economic thought reached top levels of insight in two fundamental works: Nelson (1959) on the economics of basic scientific research and his role as editor of the NBER volume (Nelson, 1962) in which – together with other classic contributions, such as Arrow (1962) – he published his magisterial study on innovation, the transistor, and the Bell Labs. In both studies, Nelson is becoming increasingly aware of the market failures surrounding innovation activities, the role of the public sector in framing the rate and direction of technological advance, and the need to consider markets as interrelated with professional associations and Universities. In another classic paper of this time (Nelson and Phelps, 1966), the authors propose a formal model to deal, precisely, with scientist training and its role in growth, a model that has permeated contemporary endogenous growth theory [as one of the strategies to deal with human capital, see Lucas (1988); Aghion and Howitt (1998)]. The Nelson–Phelps (1966) function has also been integrated into models of coevolution and catch-up by Almudi et al. (2012). It is clear that the set of works that we have mentioned above, can also be framed in the context of discussions on technology and alternative socio-economic systems during the Cold War. Note that, although the expansion of macroeconomics in the 1950s, 1960s, and 1970s (the role of fiscal vs monetary policies, Central Banks, the inflation-unemployment trade-offs) was the most popular topic in the profession (Friedman, 1968, 1970; Lucas, 1972; see also Blanchard and Fischer, 1989), it is undeniable that the strategic role of technology and related institutions (Rosenberg, 1976; Nelson, 1980, 1981), the problems of calculation in centralized systems (Hayek, 1988), and the determinants of growth in both sides of the Cold War (Abramovitz, 1956; 1986) were important for global leadership. These debates and analysis spilled over the field of growth theory and cross-fertilized with institutional and socio-political studies in public choice (Buchanan and Tullock, 1962), administrative behavior (Simon, 1957), the theory of democracy (Dahl, 1956; see recently Hodgson, 1999, 2015), and political philosophy (Sen, 1970; Rawls, 1971; recently Acemoglu and Robinson, 2019). Dick Nelson incorporated his research advances into his teaching at Yale during the 1970s and early 1980s. Thus, on the one hand, he taught industrial organization and public policy; 49

Isabel Almudi and Francisco Fatas-Villafranca

on the other hand, he used to teach the economics of technical change, while also playing a leading role at the Institute for Social and Policy Studies. From all these activities and the new insights that he was finding, Nelson was figuring out in the 1960s, 1970s, and early 1980s a view of capitalist societies as mixed, highly complex, and continuously changing systems. Even when he had to explore specific markets, Nelson was gradually abandoning the idea of market failures since, as he usually reminds us, all markets are essentially characterized by externalities, certain public good aspects, asymmetric information, the market power of some agents, incompleteness, distortions, and radical uncertainty. There are no real markets without these features. His (increasingly) multidisciplinary and mixed approach to modern societies, led him to detect that capitalist systems may undergo pathological trajectories of change. Perhaps, it is his book The Moon and The Ghetto (1977) the piece in which Nelson synthesizes, for the first time, his concerns on how the uneven evolution of human know-how in distinct but equally important realms of social activity (education, literacy, health, income distribution policies) can create enormous problems of social organization. Clearly, the combination in Nelson of his interest in technological advance as a source of growth, and his conception of capitalist change as arising from a complex network of interactions across mixed realms, explains why he never felt fully satisfied with Schumpeter’s entrepreneurial innovation theory. For Nelson, the distinctions between invention, innovation, science and technology, practice, and understanding are not so neat; of course, there are differences, but they are subtle, fuzzy, and must be carefully studied. In his opinion, technology should never be characterized as being just a body of practice, but also involving understanding (obtained in interaction with universities, public–private partnerships, and R&D labs) and being an intrinsic part of organizational change. It is a body of practice and understanding that develops through parallel but highly heterogeneous paths, favored or blocked by regulations, conditioned by ethical considerations, showing context-dependency, and entailing complicated organizational adaptations. This view on technology and innovation is more elaborated than Schumpeter’s brilliant (but preliminary) insights during the first half of the 20th century. In fact, as we will see in the next section, the Nelson–Winter proposal includes and often generalizes several Schumpeterian ideas. For instance, Schumpeter mark I and Schumpeter mark II can be reconciled as being alternative outcomes from a generalized Schumpeterian theory, depending on certain market and institutional conditions, and on sectoral technological regimes. Likewise, the Nelsonian view on the uneven paths of technological advance in distinct activities and sectors puts forward a warning regarding the potential of innovations. More precisely, Nelson highlights that not all social problems are amenable to technological fixes (a crucial point very relevant nowadays). We may be going to the moon, whereas, at the same time, in the same society, we may be ineffective in improving the way in which children learn to read or to do math. Having criteria to distinguish (ex-ante) sectoral discrepancies in this respect is essential, as we have recently shown, along the lines of Nelson (1977), in Almudi et al. (2016, 2021, and 2022) for the case of energy storage. Finally, note that as Richard Nelson increasingly claimed during those years that technologies develop at highly uneven rates, in context-dependent manners, entailing organizational adaptations, and with variations depending on sectors, he was becoming close to Sid Winter’s evolutionary conceptions. The combination of both critical views was explosive. In fact, as we know today, the Nelson–Winter collaboration in the 1970s and 1980s (Nelson and Winter, 1973, 1974, 1977, 1978, 1982), the work along these lines by some of their – then – students such as Malerba (1985), their interactions with scholars at SPRU (Freeman, 1982, 1987; Dosi, 1982, 50

The foundational evolutionary traverse of Nelson and Winter

1984, 1988; Pavitt, 1984) and with economic historians at Stanford (Rosenberg, 1982), all this led to a disruptive departure from mainstream economics. This departure established, with the passing of time, the seeds and roots of the new school of evolutionary economics. As an example of these disruptive advances, Nelson and Winter (1982) proposed a new methodology to do economic science, one that cannot be easily reconciled with mainstream practices. They distinguished between what they called appreciative and formal theorizing, with the former being close to empirical studies, focused on finding the key variables and mechanisms that might be going on in reality, and the latter (formalization) devoted to checking the consistency of the theories, sharpening the arguments, and highlighting new ways. This methodology is, probably, not fully shared as the unique heuristic by all evolutionary economists, but it has been enormously influential and fruitful up to our days [see the history-friendly models by Malerba et al. (2016)]. In the next section, after having explored the context and early works of our protagonists, and once we have arrived at the originating sources of evolutionary economics, we are going to analyze the Nelson and Winter (1982) foundational book, An Evolutionary Theory of Economic Change.

3.4

The foundational evolutionary traverse

As we have seen, the foundational evolutionary traverse of Nelson and Winter had begun, at least, fifteen years before the publication of their 1982 book. But the content of this book conveys and sharpens many of the individual and common insights that our protagonists had devised during even more time. Considering the limits of this chapter, we would like to put shortly that, regarding its contents, Nelson and Winter (1982) deserves special recognition for at least three reasons (some of them fully explicit in the text, while others being implicit but ready to inject an overhaul in economics): first, there is a sharp critical attack on why economic change is deficiently addressed by mainstream economics; second, they propose consistent parts of a new confronting theoretical vision to study change; finally, Nelson and Winter suggest avenues for further research to come up with an alternative paradigm in economic science. In this ambitious program, the book is much in line with another classic, Dosi et al. (1988), and with the Freeman project at SPRU (Freeman, 1982, 1987, 1990), a program that is developing today along the lines of Dosi (2000), Dosi and Nuvolari (2020), and Dosi (2023). In setting the stage for their critique, remind from the previous sections that Nelson and Winter had already come to see economic change in a broad manner. They had in mind at that time, not only the conventionally accepted long-run transformations in modern economies entailing income and consumption per capita growth, capital accumulation, and productivity enhancing technical advances, but also (and primarily) the ongoing organizational, multisectoral, techno-scientific, and institutional dynamics that make those quantitative trends possible. More precisely, Nelson and Winter (1982) state that firm adaptations, and endogenous novelties in organizations and technologies, always shape and, in turn, are re-shaped in real time, by scattered interactions with rivals and institutions. These processes drive industrial dynamics and macroeconomic change. In a neat Schumpeterian spirit, they point out technological change and the needed institutional coadaptations as the crucial driving forces in all economic dimensions. From this perspective, Nelson–Winter proceed to present their demolishing critique to mainstream economics (GET, neo-Keynesian models, the successive neoclassical neoKeynesian synthesis). They show why these approaches cannot explain correctly the sources 51

Isabel Almudi and Francisco Fatas-Villafranca

of economic change and, therefore, most economic problems, since for Nelson and Winter economic systems are always in motion. Their attack points to certain core assumptions that standard theories were carrying on since the marginal subjectivist revolution and the first quarter of the 20th century. Thus, they argue against supply-demand equilibrium settings, with fully (or probabilistically) informed maximizing agents facing financial and technostatic constraints, with the agents simply reacting to prices (or expected prices) to devise their plans, and these plans being compatible (or becoming compatible in transient paths) in stable (static or trendy) stationary states. This is the core of standard theory that Nelson–Winter denounce as being incompatible with a realistic analysis of economic problems. Let us briefly assess the scope of this broadly anticipated critique, and then we will go to the details of the critical arguments and their possible solutions. In our opinion, since the Nelson–Winter critique goes against the fully rational (marginalist) characterization of agents in modern societies, with these societies being primarily perceived as market systems in equilibrium (in different formal concretions), it turns out that the Nelson and Winter critique is extensible to most of the interwar developments that we have posed in Section 3.1, and their post WWII realizations, all the way to the 1980s. This range of works would include the following: comparative static exercises and standard dynamic system analysis in neoclassical models; the orthodox Keynesian implementations; topological (algebraic or differential) GET-models and extensions; many industrial organization studies based on classic Game Theory or in (old-style) imperfect competition models; the successive vintages of dynamic general equilibrium settings with intertemporal optimization (deterministic, or with rational expectations á la Muth (1961), with monetary or real impulses, focusing on cycles or dealing with growth). Hence, the critique attacks the stream of (micro and macro) orthodox advances until the 1980s (Arrow and Hahn, 1971; but also Debreu, 1970, Dierker, 1974, Mas-Colell 1985; Lucas and Prescott, 1971, Kydland and Prescott, 1982, Lucas, 1983 and Sargent, 1987; IO advances such as Dixit and Stiglitz, 1977, Dasgupta and Stiglitz, 1980, or Tirole, 1988; and Neo-Keynesian settings compiled in Blanchard and Fischer, 1989). These approaches, either static or dynamic, deterministic or stochastic, always involve sufficient conditions for some sort of stationary state to exist (and often even to be unique), and include compactification and convexification possibilities, with continuity properties for functions and correspondences, in such a way that fixed point arguments can be used in the proofs; they often incorporate random shocks with nice properties (null mean, identically-independently distributed with low constant variance, low autocorrelation orders and stationarity, even Gaussian shapes) for quick return to the trend to be assured; and in the most sophisticated cases (in which measure theory and contractions apply) they involve stochastic processes verifying sufficient conditions for convergence (or almost convergence) in order to focus the analysis on some kind of “state of rest” (attractive stochastic trends, or well-behaved limit-distributions). This is the scope of the critique that, in our opinion, is explicit (or sometimes simply implied) by the Nelson–Winter attack. Now let us explain the critical arguments in some detail. In Nelson and Winter (1982), the critique is especially powerful regarding organization theory and the standard theory of the firm. As Nelson and Winter argue, for mainstream theory, firms are quasi-identical agents that present neat objectives (maximizing profits drawing on consistently evaluated alternative production plans, or max-profits defined over calculable alternative flows of accumulated expected returns), constrained by traceable and accessible choice sets (closed, convex and “boundable” production sets, identified with the state of technical knowledge), and subject to “visible” vectors of market prices (or probabilistic 52

The foundational evolutionary traverse of Nelson and Winter

sequences of expected prices). In case that changes in those aspects were to occur, they would be either exogenous for firms or (if firms had market power) controllable by them with clear results. With these firms facing standard demand-side settings, prices (or sequences of “rationally expected” prices) would allow for optimal action, and the whole system (industry or the economy as a whole) will be (or will move without chronic problems) to a situation of mutual-compatibility of plans; that is to say, sustained stationary states in general equilibrium. This characterization of agents (with or without externalities, with the possibility of other standard market failures or rigidities, transitory rationings, in fully known strategic settings in the case of Games) is what, essentially, allows firms to rationally calculate and detect their optimal choices (given their objectives and constraints and those of other agents) and behave accordingly, with instantaneous (or dynamically assured) overall compatibility of actions. As Nelson–Winter argue, it is this core framework that rules out the possibility for standard models to deal with real economic transformations and, in turn, to cope with the economic problems of systems in motion. As they say, in analyzing mainstream models, one can simply check how equilibria (or equilibrium paths) are modified depending on policy interventions, exogenous variables, parameters, or random shocks, but always being sure that the system will recover some kind of stationary state. The assumption of reversibility of choices is implicit in these settings, as firms can move smoothly across the choice set in their search for optimal options. Scalability, perfect replication, high elasticity of input substitution, free disposal of input-excesses, non-costly ceases of activity, and easy information acquisition are all typical complementary assumptions. The orthodoxy rules out irreversible mistakes, radical uncertainty, discontinuities, non-convexities, and disconnections in choice sets or discrete spaces. And, if bad choices were made, market processes would always reconduct the system towards optimal (or tractable sub-optimal) states of rest, in a renewed appeal to Friedman (1953) defensive argument. By the 1980s, in this mainstream context, economic growth was explained (even in the stochastic models) by defining dynamic general equilibrium conditions with exogenous population growth, in which maximizing firms managed to expand their use of inputs (capital, labor, financial assets), increasing capital intensity, and taking advantage of exogenous technological changes that entailed smooth and traceable modifications in the production possibility set; all these elements led to output growth, consumption growth and increasing employment levels (and the corresponding variables per capita) in the long run [from Solow (1956, 1957), and Ramsey-Koopmans-David Cass (1965), to the overlapping generations models of Diamond (1965), and including the New Classical Macroeconomics (Lucas, 1983; Sargent, 1987; Hall, 1990) and the RBC-DSGE models in Stokey et al. (1989)]. It is noticeable that, the hypothesis that firms may choose among a wide range of technological options (input-output combinations) including those never used before, or those located far away from the previous technical state, is assumed by ruling-out transition failures, knowledge gaps, organizational adaptations, and problems of path dependency and lock-in (Dosi, 1997; Hodgson, 2001; Dopfer and Potts, 2008; Dosi and Grazzi, 2010; Metcalfe, 2010). This mainstream lacuna will instill interesting debates between evolutionary and neo-Austrian scholars in the 1990s (Kirzner, 1992; Loasby, 1999; Potts, 2000). Notwithstanding its value, we would like to mention that the Nelson–Winter critique in the 1980s had shortcomings that were to be improved in the following decades. For instance, they left aside consumer behavior, the critique did not go deep into sophisticated financial and monetary aspects, and it was almost blind to government budget dynamics and aggregate demand. In any case, the diagnosis in Nelson and Winter (1982) is clear: orthodox 53

Isabel Almudi and Francisco Fatas-Villafranca

theory is not suitable to explain how real firms engender and react to changes in technologydriven market conditions and how these innovations affect industry structures and economic growth. An alternative formulation from a Schumpeterian evolutionary perspective should consider that innovations (and their consequences) come from more realistic and elaborated corporate and organizational models combined with domain-specific technical, institutional, and non-market factors, precisely those factors that our protagonists had been studying for years (Sections 3.2 and 3.3 above). The inadequacy of orthodoxy to deal with economic change leads Nelson and Winter to put forward some consistent pieces for a new framework, an implicit overall challenge to mainstream economics, and a rich research agenda for the following decades. We want to briefly mention several elements. Firstly, organizations are seen in Nelson and Winter (1982) as heterogeneous agents (vs orthodox representative agents) endowed with different and limited operational capabilities and organizational routines, pursuing idiosyncratic objectives (profits, sales, market shares, growth rates, sharing dividends, permanence) in a boundedly rational way (Simon, 1983). Organizational reactions to environmental changes are local and based on specific, practical, often tacit, and context-dependent previous knowledge that constraints firm-specific action. These knowledge features are encapsulated – and carried on by organizations – into collective operational routines. Routines are seen as the key (internal) knowledge coordination and operational mechanisms that drive organizational responses to the information received from outside, and from within. Bounded-rationality of profit-seeking firms (vs fully rational orthodox behavior explained above) means that they try to do better (instead of looking for the best option), and they adapt accordingly, along the lines of the satisficing behavior hypothesis (maintaining routines, or patterns that have worked well, unless they observe performance levels below specific thresholds). In this context, innovations are responses to specific problems or environmental changes; the search for these responses is costly, local, and radically uncertain. R&D activities are carried out and oriented from preexisting operational routines/technologies (local search), and they are generally focused on firm-specific goals, displaying discrete trajectories (involving topological complications that do not appear in the mainstream) everything within self-organizing firm populations. Secondly, the degree of innovation success and how it affects the firm relative presence in markets depends not only on inner efforts, but also on rival behaviors (incumbents or new entrants), on the base of technological opportunities in the sector/industry, the degree of cumulativeness in knowledge production, and the extent to which the new discoveries can be privately appropriated by firms (patents, advantages of pioneers, entry barriers). These sectoral parameters would allow the analyst to simulate alternative technological regimes (Winter, 1984) and industrial trajectories, which may resemble Schumpeter mark I or mark II, depending on alternative simulation settings. Considering all these aspects, major innovations (radical innovations) may happen but with a much lower probability than minor ones (incremental innovations), and often (not always) appear incorporated in new entrants. Apart from their inner innovations, firms can opt for imitating the technical solutions of their competitors. But in Nelson and Winter (1982), imitation (of techniques or routines) not always works, since cross-firm replication through imitation is perceived as imperfect, and involves the adoption of operational routines that (often) only function within very specific contexts. Hence, imitation is also uncertain, costly, and not very different from pure innovation. This is a highly original insight which breaks with the standard conception of technological knowledge as a quasi-public good. More recently, replication 54

The foundational evolutionary traverse of Nelson and Winter

dynamics and routines have been largely explored in Winter (1995, 2005, 2006), Becker (2010), Hodgson and Knudsen (2010), and the sectoral framing aspects have received attention in Dosi (1988, 1997), Winter et al. (2003, 2007), Dosi et al. (2005), Dosi and Nelson (2010), and Almudi et al. (2013). Remarkably, the evolutionary vision presented above has been clearly differentiated from the behavioral theories by Kahneman (2003), Richard Thaler and colleagues, in that evolutionary economic theory does not focus on framing psychological distortions, noise, or in paradoxical perception-decision phenomena, but in bounded-rationality, as defined in Simon (1983) (see Nelson, 2018). Finally, in the models devised for industry studies or for growth theory, Nelson and Winter propose studying populations of boundedly rational diverse (profit-seeking) firms, with market selection operating at the organizational level, so that, more successful organizations (embodying better-fitted routines) gain presence (in terms of profit-driven growth, market share), while others lose it. The frequency distributions (related to firms and specific traits) change through market selection, entry-exit, and the imperfect replication of successful traits. Since firms need to adapt to changing contexts, they may try to imitate the routines of the most successful organizations (which are always subjectively and imperfective perceived by imitators), or try to inner-innovate to adapt. At any time, the search for new options at the firm level is always subject to radical uncertainty, and organizations may or may not fail when pursuing their objectives. As a result of this, the non-equilibrium competitive processes modeled in evolutionary economics endogenously generate (parameter-dependent) highly complex dynamics for market concentration, prices, R&D spending, and sectoral growth. These models often have an algorithmic implementation with the subsequent computational exploration; the specific forms in Nelson and Winter (1982) are simply concrete stylized prototypes to be developed. In fact, the original models were very soon extended to study new phenomena in industrial dynamics by Winter (1984), Rothblum and Winter (1985), Silverberg et al. (1988), Metcalfe (1994, 1998), and Dosi et al. (1995). In the case of evolutionary growth models, departing from the simple exemplar in Nelson and Winter (1982), new models appeared soon, involving multiple sectors (interlinked through vertical and horizontal interactions), with the whole economic system displaying structural change (changing relative importance of sectors in the aggregate), with the possibility of new sectors entering, others declining, and the aggregate being depicted as a restless system composed of co-determined subsystems. This was the picture that Schumpeter (preliminary) envisioned, and that the evolutionary authors of the 1990s and 2000s clearly established (Chiaromonte and Dosi, 1993; Dosi et al., 1994; Silverberg and Verspagen, 1994; Saviotti, 1996; Saviotti and Pyka, 2004; Metcalfe et al., 2006; Fatas-Villafranca et al., 2008, 2009, 2012; Almudi et al., 2013). In modeling these evolutionary processes [and this includes the models in Nelson and Winter (1982)], the mathematics of evolution that had been developed since the 1930s, and the theory of stochastic processes (including the time-series econometrics of the 1980s and 1990s) were essential (see Andersen, 1996; Valente and Andersen, 2002; and the wide comparative studies in Silverberg and Soete, 1994; Dosi, 2000, and Metcalfe et al., 2006). The technical tools used in evolutionary economics have been enriched since the 1980s by using evolutionary models of population games, computational methods (ABMs), and stochastic dynamic networks (Almudi and Fatas-Villafranca, 2021). We will talk about these developments later on. Of course, there are more themes in the 1982 book. But, for brevity, and considering the scope of this book, in the next sections, we will follow the development of these ideas, since they delimit the core of contemporary evolutionary economics. 55

Isabel Almudi and Francisco Fatas-Villafranca

3.5

The golden decades

The decades during the 1990s and 2000s saw an explosion of research in evolutionary economics (Foster and Metcalfe, 2001; Witt, 2003; Dopfer, 2005; Hanusch and Pyka, 2007). This is the reason why we call this period, the golden decades. With the advantage of hindsight, we can say that Nelson and Winter (1982), Dosi et al. (1988) and from a slightly different perspective Anderson et al. (1988), all gave critical momentum to many developments that were advancing along similar lines (Arthur et al., 1997; Weibull, 1995; Dosi, 2000; Dopfer and Potts, 2008; Hodgson, 2019). This is a fascinating phase in which, with Nelson and Winter developing their views separately but benefiting from their joint foundational traverse, and with cascades of controversies around epistemological and historical bases, new tools, firm theory, the notion of rationality, industrial dynamics, economic growth and social organization, technical change, and policy-making, the evolutionary economics paradigm took shape. In this section, as in the whole chapter, we will focus on the role of Nelson and Winter, and we will be brief because the other chapters in the handbook cover the specific developments. Throughout his section, we will delineate (just as an outline) what Winter (2014) has called the consolidated and empirically well-supported “beachhead” in evolutionary economics. That is, a strong body of theoretical, instrumental, and applied evolutionary knowledge that was developed during the 1990s and 2000s and that stands well up to empirical testing. As we will see in the next section (the final one), recent historical traumatic events and the accumulation of insights during the golden decades have produced a new explosion of innovative work in evolutionary economics in the 2010s and early 2020s. These advances have moved the consolidated frontier far beyond Sid Winter’s (2014) beachhead. But before we move to the front, let us study very briefly the golden decades. After having served for some years at the US General Accounting Office as chief economist (Winter, 1992), Sidney Winter moved to Wharton and developed his previous contributions (in coming up with a new theory of the firm) in line with certain areas of organization science and management studies. More precisely, as Teece et al. (1997) had been proposing and studying in detail, the dynamic capabilities of specific firms might be considered key factors in explaining the origins of competitive advantage and, of course, the ultimate sources of innovation and growth (see the specific chapter in this book). Although orthodox scholars in the field of strategy remained a bit skeptical about the role of this conception, evolutionary economists [Dosi et al. (2000) and Winter (2003)], participated decisively in the debates, and supported the notion of “dynamic capabilities” by connecting it with the more basic concept of organizational capability, all the way down to the original evolutionary notion of organizational routine. They also provided industrial models in this regard (Zollo and Winter, 2002). The evolutionary argument and, clearly, that of Sid Winter basically states that a firm capability (in general) may be seen as a high-level set of routines that, jointly with the implied input flows, provided firm’s management with options to decide how to produce specific outputs. He uses the notion of routine (as we have already seen in previous sections) as denoting a pattern of collective behavior that has been learned by operational repetition, has proved useful and operational, and that involves (partly) pieces of tacit and non-articulable knowledge. Thus, a bundle of interlinked and connected routines would provide the carrying organization with idiosyncratic operational capabilities, and by adding the qualifier “dynamic”, one may be pointing out to capabilities related to innovation and change. Then, dynamic capabilities would be the specific type of

56

The foundational evolutionary traverse of Nelson and Winter

capabilities that are essential for managing deliberate organizational changes and innovative adaptations. As it is well known, there is a whole field of research along these lines, that is further studied in other chapters, and which can be seen in the contributions to Nelson (2018). It is straightforward that these new insights regarding organizational innovations are related to the very ultimate sources of economic growth. This was the theme in which Dick Nelson played a leading role during this phase. Richard Nelson moved to Columbia University and, with time, became director of the Program on Science, Technology and Global Development at the Columbia Earth Institute. One of the debates in which Nelson participated vividly during this period, regarded the suitability of the new endogenous growth models to explain economic change. More precisely, Nelson (1998) discussed these new models in the light of the nascent evolutionary theory of economic growth. It was clear in the 1990s that for both, the mainstream models and evolutionary theories, the proposition that technological advance is the source of longrun productivity growth was firmly established. Although the rates of technological progress may differ enormously among industries, highly innovative sectors show high R&D intensity, large innovations in supplying industries, or both. Regarding these basic statements, there was agreement among both schools of thought. But, whereas mainstream economists approached growth dynamics by coming up with neoclassical endogenous growth settings (Lucas, 1988; Romer, 1990; Grossman and Helpman, 1991; Aghion and Howitt, 1998; Acemoglu, 2008; Benassy, 2011; Romer, 2012), evolutionary theorists characterized growth as an out-of-equilibrium process necessarily involving structural change, displaying overlapping fluctuations of distinct frequencies, being driven by the radically uncertain creation-destruction of firms and sectors, and generating problems of income distribution and institutional co-adaptations (Chiaromonte and Dosi, 1993; Silverberg and Verspagen, 1994, 2005; Silverberg and Soete, 1994; Nelson, 1996, 2005; Saviotti and Pyka, 2004, 2013; Metcalfe et al., 2006; Winter, 2008; Fatas-Villafranca et al., 2012; Valente et al., 2015). As Nelson (1998) explained, the differences between both lines of analysis (endogenous growth theory, and the evolutionary stream) are extremely significant. In the evolutionary literature, nothing resembling the dynamic optimization plus general equilibrium approach (even with imperfect competition and stochastic extensions) that we see in endogenous growth models can be considered an acceptable basis to analyze economic growth. Evolutionary economists also consider some degree of market power, endogenous firm-R&D investments, human capital provision, and the emergence of increasing returns at a macro-level as key drivers of growth. But, in contrast with the equilibrium settings from which neoclassical authors deduce “quality ladder properties” or “stealing effects” (Grossman and Helpman, 1991; Aghion and Howitt, 1998), the evolutionary scholars argue that growth emerges from the scattered innovative action of bounded-rational profitseeking firms, which are highly diverse, compete in changing frames in which multiple equilibria may exist, and non-linear dynamics (shocked by entry-exit of firms and sectors) always produce complex dynamics. As a brief comment, it is obvious that for evolutionary scholars, the so-called transversality conditions that accompany the principle of maximum in optimal control (with the use of Hamiltonians) or the Euler first-order marginalistintertemporal equations that underlie the dynamics of control variables, not to say the perfect knowledge of dynamic stock-flow state equations or the permanent market-clearing conditions that formally close dynamic optimization models, all this conveys a distorting characterization of economic dynamics. Note that the so-called transversality conditions rule out Ponzi games, and add in these models to assure that fully rational agents locate the 57

Isabel Almudi and Francisco Fatas-Villafranca

(unique and null-measure) convergent path towards the growth steady state; and they do it in saddle-path-type instability models, something that even from a rough mathematical point of view is a bit intriguing. These comments are enough to discard endogenous growth theory as a valid explanation of change from an evolutionary perspective, when it is increasingly clear (from an empirical viewpoint), that technology advances by transforming the whole context in which firms operate, and entails the appearance of radical “unknowns”. Innovation (by definition) places us within the realm of what has never been tried before, a realm that is formed by objects, processes, pieces of information, institutional arrangements, and potential human reactions of an unprecedented nature (Arrow, 1994; Witt, 2009, 2014). Besides this, as Metcalfe (2001) and Nelson (2008) strongly emphasized during those years, economic growth and technological advance cannot be understood without considering the contribution of supporting institutions during the process. Nelson (2018) has been insisting for decades that the role of science and Universities in economic change is more complex than the characterization they receive in endogenous growth theory (compare Lucas, 1988, with Almudi et al., 2012, 2021). In Nelson’s studies, scientific research in fields like electrical or chemical engineering, computer science, or pharmaceutical biotechnology coevolve with practical and profit-oriented corporate developments in non-trivial ways. Likewise, the role of patents, multiple and highly heterogeneous efforts going on at the technology frontier, driven by very specific corporations in distinct sectors, all interlinked with concrete university systems, regulatory agencies, legal frames, and professional bodies, are key explanatory factors of technological innovation and growth. These mechanisms are essentially distorted (and often erased in their rich intricacies) in endogenous growth models (Nelson, 2005, Dosi and Nelson, 2010). Regarding these aspects, Nelson and Winter (2002) state that the study of sectoral processes of Schumpeterian competition, the empirical study of specific innovative sectors, and looking for the intersectoral connections to understand growth, is necessary but not sufficient. Here we find (as the necessary complementary piece) one of the most significant insights of the golden decades, one that had been already explored by Chris Freeman and colleagues, but that becomes very much improved and generalized during this period: we refer to the observed coevolution of institutions, firms, technologies and market mechanisms, interlaced in complex multi-agent dynamic networks, that shape the so-called national (and even sectoral) systems of innovation (Lundvall, 1992; Nelson, 1993; Malerba, 2002). The concept of system of innovation has been playing a central role since the 1990s in evolutionary economics, and it has proved useful to understand: the sources of industrial leadership in high-tech sectors (Mowery and Nelson, 1999; Malerba, 2002; Murmann, 2003; Fatas-Villafranca et al., 2008; Almudi et al., 2012); the global processes of economic development understood as global learning processes (Verspagen, 1991; Malerba and Nelson, 2013; Almudi and Fatas-Villafranca, 2018, 2021); the role of the state in evolutionary economic theory (Dosi, 1995; Mazzucato, 2013); and the evolutionary foundations of economic geography (Boschma and Martin, 2010). There is nothing in mainstream microeconomics or macroeconomics that may accommodate the analysis of these technoinstitutional, corporate, and market networks [nothing, from recent developments in Game Theory as in Fudenberg and Tirole (2000), or in advanced GET (Balasko, 2016), to the recursive macroeconomic theory with Bellman equations and stochastic dynamic programing in Ljungqvist and Sargent (2018) or Miao (2022)]. Nevertheless, new tools have been devised regarding network theory (Jackson, 2008), or in the modeling of learning 58

The foundational evolutionary traverse of Nelson and Winter

processes in evolutionary frames (Samuelson, 1997; Sandholm, 2010; Hart and Mas-Colell, 2013) that can be of technical help for evolutionary economic theory. Then, for brevity and given the limitations of space, we can sum up the results of this section by saying that, with Nelson and Winter being highly active in the debates during the 1990s and 2000s, a compact and well-grounded body of appreciative and formal theoretical work was obtained regarding industrial dynamics and growth, together with advances in organization theory, economic development, innovation systems, evolutionary economic geography, and policy studies. These results stood perfectly up to empirical scrutiny (Dosi et al., 2017) and they constitute what Winter (2014) named the “consolidated beachhead” in evolutionary economics. In the next section, we see how drawing on these established foundations, and stimulated by a sequence of traumatic events that have shocked the world during the last decade, evolutionary economics have developed new insights during the 2010s up to our days. We will finish the chapter by showing how contemporary evolutionary economics has moved well beyond the consolidated beachhead.

3.6

Far beyond the consolidated beachhead

The turbulent decade (2008–2022) has seen a sequence of shocks and traumatic events that have dismantled ideas and structures that had been taken for granted (perhaps for too long) at a worldwide level. For the sake of brevity, we are going to simply mention recent developments in evolutionary economics fostered by the Great Recession (2008–2016), the COVID-19 pandemic with the great lockdowns that almost paralyzed the world (2020–2022), and the increasing geopolitical tensions that lead us towards a multipolar dangerous world. It is remarkable how these apparently independent events have uncovered hidden weaknesses; silent fractures that were stalking our dream of a global liberaldemocratic (capitalistic) order, a state of things that some had considered as being approaching the end of history (Fukuyama, 1992). These events have been extremely traumatic, and, together with the strong bases on which evolutionary economics had been left after the golden decades, they have inspired new frames, developments, and results that have expanded the frontier of evolutionary economics. The Great Recession revealed the dark side of global financial and trade generalized interconnections, a network of transnational interactions that was relying on the belief that efficient market-allocations did not need tight regulations, and in which financial innovations (audited by agencies “capable” of calculating increasing risks) allowed to cover all sorts of funding needs (Caballero, 2010; Eggertsson and Krugman, 2012). When investment banking went into bankruptcy -leading the global economy to its most severe crisis since the 1930s – the world woke up from its market dream, and the fears regarding secular stagnation, liquidity traps, increasingly unequal income distributions, high unskilled unemployment in Western societies and socio-political revolts came to stay. Many of the fatal conceits underlying the crisis, and the huge shortcomings of mainstream NK-DSGE models in vogue at that time (Woodford, 2003) were denounced by Sidney Winter (2010) in his testimony submitted to the US Congress. The reaction of evolutionary economics to this situation involved important breakthroughs from the consolidated beachhead in, at least, three directions: i) the most influential advance has been the key “Schumpeter-meeting-Keynes” generation of evolutionary macro-models put forward by Dosi et al. (2010) and Dosi et al. (2013). These models are complementary to the complexity approaches in Gallegati et al. (2017) and Wilson and Kirman (2016). This overall new approach has been discussed and studied in detail 59

Isabel Almudi and Francisco Fatas-Villafranca

by Dosi and Roventini (2019); ii) there is a second essential body of evolutionary macromodels that recover post-Keynesian ideas and institutionalist notions, and incorporate them in evolutionary settings which display demand-driven structural change. We find them in the Ciarli et al. (2019) or Valente et al. (2015) version, and/or in replicator models with vertical integration and value chains in Cantner et al. (2019); iii) there is also a body of stylized (formally tractable) evolutionary macro-models (see Fatas-Villafranca et al., 2012) that tackle, firstly, the dynamics of income distribution and unemployment in relation with growth, technological change and overlapping cycles of distinct frequencies. Secondly, in a variant presented in Almudi et al. (2020) in Metroeconomica, these models include the role of the banking sector in innovative economies with debt, in which inflationary paths lead to rising interest rates, and this mechanism may slowdown innovation, deteriorate income distribution and engender “big rips” in economic growth. The three lines of advance i), ii), and iii) have moved evolutionary macroeconomics into unexplored territories which, apart from reacting to the turbulent decade (2008–2022), have also tried to deal with concerns expressed by evolutionary economists of previous generations (Witt, 2014; Winter, 2017). Somehow related to this, there have been significant recent advances related to the evolutionary theory of consumer behavior and demand. This is a shortcoming that we have already mentioned in the chapter. Ulrich Witt (2001) began to face this issue two decades ago, and perhaps stimulated by the ideas in Nelson (2013, 2018) this line is fructifying in new models and studies by Chai and Baum (2019), Fatas-Villafranca et al. (2019, 2007), Almudi et al. (2013) and Valente (2012). These new approaches to consumer behavior have appeared, in parallel, with a more generalized interest in the cognitive and intentional characterization of agents in evolutionary theories [a classic issue in Dopfer (2005) that has benefited from insights in Nelson (2016) and Muñoz et al. (2011)]. Quite apart from this, the COVID pandemic and the great lockdowns with distribution bottlenecks that followed have shown that, when economic systems collapse (and this is a very real possibility in complex systems), getting things in motion again is not a trivial task. Thus, in two complementary works that have recently appeared, Jason Potts and colleagues (Allen et al., 2020) and Almudi and Fatas-Villafranca (2021) have analyzed co-evolving systems which display potential virtuous paths vs possible collapsing trajectories and have looked for alternative ways to re-invigorate the systems in post-traumatic conditions. Very much in line with former ideas by Chris Freeman (2019) and by Dick Nelson on interrelated selection processes (Dopfer, 2005; Dosi and Nuvolari, 2020) the recent contributions on coevolution consider explicitly (and in alternative formalizations) the mixed dynamic and co-determined nature of modern economies. Drawing on alternative co-evolutionary models (that combine replicator dynamics, ABMs, and networks) Almudi and FatasVillafranca (2021) detect catalyzing and blocking factors which may lead either to sustainable paths or to trajectories of collapse. This stream of works is also in line with Andreas Pyka (2017) new concept of “dedicated innovation systems”, and with the overall characterization of capitalist economies as mixed evolving systems in Nelson (2022). Finally, the geopolitical tensions and socio-political turmoil that we see in most societies after the turbulent decade remind us that not all social problems are amenable to technological, market, and economic policy-driven fixes (the classic theme in Nelson, 1977; formalized in Almudi and Fatas-Villafranca, 2022). We need to reconsider aspects of political economy from an evolutionary perspective (Fatas-Villafranca, 2011; Almudi et al., 2017). The rise of populism and increasing inequalities in the dangerous multipolar world that comes, require that we deal with the dynamics of power (hard political power in the classic 60

The foundational evolutionary traverse of Nelson and Winter

sense), very much along the lines of Dosi (1995) and Dosi et al. (2020). This would be a new step in recognizing, as Nelson (2022) does, that in all human systems in history, non-market aspects are more deciding factors for social life than isolated market conditions.

Acknowledgments This research has been funded by projects PID2019-106822RB-I00, S40 _20Rand S40_23R.

References Abramovitz M (1956) Resources and output trends in the United States since 1870. American Economic Review, 46, 5–23. Abramovitz M (1986) Catching up, forging ahead and falling behind. Journal of Economics History, 66, 385–406. Acemoglu D (2008) Introduction to Modern Economic Growth. Princeton University Press. Princeton. Acemoglu D and Robinson J (2019) The Narrow Corridor: States, Societies and the Fate of Liberty. Penguin. London. Aghion P and Howitt P (1998) Endogenous Growth Theory. MIT Press. Cambridge, MA. Alchian A (1950) Uncertainty, evolution and economic theory. Journal of Political Economy, 58, 211–222. Alchian A (1955) The rate of interest, Fisher’s rate of return over cost and Keynes’ internal rate of return. American Economic Review, 45, 938–942. Alchian A and Demsetz H (1972) Production, information costs and economic organization. American Economic Review, 62, 777–795. Allen D, Bergh C, Davidson S, Lane A and Potts J (2020) Unfreeze. The American Insitute of Economic Research. Great Barrington, MA. Allen RGD (1960) Mathematical Economics. (2nd ed). MacMillan. London. Almudi I and Fatas-Villafranca F (2018) Promotion and co-evolutionary dynamics in contemporary capitalism. Journal of Economic Issues, 52, 80–102. Almudi I and Fatas-Villafranca F (2021) Coevolution in Economic Systems. Cambridge University Press. Cambridge. Almudi I and Fatas-Villafranca F (2022) Sarewitz-Nelson rules (for innovative choice and technology policy). In Antonelli C (ed) Elgar Encyclopedia on the Economics of Knowledge and Innovation. Edward Elgar. Cheltenham. Almudi I, Fatas-Villafranca F, Fernandez C, Potts J and Vazquez FJ (2020) Absorptive capacity in a two-sector neo-schumpeterian model: A new role for innovation policy. Industrial and Corporate Change, 29, 507–531. Almudi I, Fatas-Villafranca F and Izquierdo LR (2012) Innovation, catch-up and leadership in science-based industries. Industrial and Corporate Change, 21, 345–375. Almudi I, Fatas-Villafranca F and Izquierdo LR (2013) Industry dynamics, technological regimes and the role of demand. Journal of Evolutionary Economics, 23, 1073–1098. Almudi I, Fatas-Villafranca F, Izquierdo L and Potts J (2017) The economics of utopia: A coevolutionary model of ideas, citizenship and socio-political change. Journal of Evolutionary Economics, 22, 629–662. Almudi I, Fatas-Villafranca F, Jarne G and Sanchez J (2020) An evolutionary growth model with banking activity. Metroeconomica, 71, 392–430. Almudi I, Fatas-Villafranca F and Sanchez J (2016) A formal discussion of the Sarewitz-Nelson rules. Economics of Innovation and New Technology, 25, 714–730. Andersen ES (1996) Evolutionary Economics. Post-Schumpeterian Contributions. Pinter. London. Anderson P, Arrow K and Pines D (eds) (1988) The Economy as an Evolving Complex System. Addison-Wesley Longman. Redwood City, CA. Arrow K (1951) Social Choice and Individual Values. John Wiley & Sons. New York, NY. Arrow K (1962) Economic welfare and the allocation of resources for invention. In Nelson R (ed) The Rate and Direction of Inventive Activity. Princeton University Press. Princeton, NJ.

61

Isabel Almudi and Francisco Fatas-Villafranca Arrow K (1974) The Limits of Organization. Norton. New York, NY. Arrow K (1994) The production and distribution of knowledge. In Silverberg G and Soete L (eds) The Economics of Growth and Technical Change. Edward Elgar. Cheltenham. Arrow K and Debreu G (1954) Existence of an equilibrium for a competitive economy. Econometrica, 22, 265–290. Arrow K and Hahn F (1971) General Competitive Analysis. Holden-Day. San Francisco, CA. Arthur B, Durlauf S and Lane D (eds) ( 1997) The Economy as an Evolving Complex System II. Addison-Wesley Longman. Redwood City, CA. Balasko Y (2016) Foundations of the Theory of General Equilibrium. (2nd ed). World Scientific. London. Becker MC (ed) (2010) Handbook of Organizational Routines. Edward Elgar. Cheltenham. Benassy JP (2011) Macroeconomic Theory. Oxford University Press. New York, NY. Blanchard OJ and Fischer S (1989) Lectures on Macroeconomics. MIT Press. Cambridge, MA. Blinder A and Solow R (1973) Does fiscal policy matter? Journal of Public Economics, 2, 319–337. Boschma R and Martin R (eds) ( 2010) The Handbook of Evolutionary Economic Geography. Edward Elgar. Cheltenham. Buchanan J and Tullock G (1962) The Calculus of Consent. University of Michigan Press. Ann Arbor, MI. Caballero R (2010) Macroeconomics after the crisis: Time to deal with the pretense of knowledge syndrome. Journal of Economic Perspectives, 24, 85–102. Cantner U, Savin I and Vannuccini S (2019) Replicator dynamics in value chains: Explaining some puzzles of market selection. Industrial and Corporate Change, 28, 589–611. Cass D (1965) Optimum growth in an aggregative model of capital accumulation. Review of Economic Studies, 32, 233–240. Chai A and Baum C (2019) Demand, Complexity and Long-Run Economic Evolution. Springer Nature. Cham. Chiaromonte F and Dosi G (1993) Heterogeneity, competition and macroeconomic dynamics. Structural Change and Economic Dynamics, 4, 39–63. Ciarli T, Lorentz A, Valente M and Savona M (2019) Structural changes and growth regimes. Journal of Evolutionary Economics, 29, 119–176. Coase RH (1937) The nature of the firm. Economica, 4, 386–405. Coase RH (1960) The problem of social cost. Journal of Law and Economics, 3, 1–44. Cyert R and March J (1963) A Behavioral Theory of the Firm. Prentice Hall. New York, NY. Dahl R (1956) A Preface to Democratic Theory. University of Chicago Press. Chicago, IL. Dasgupta P and Stiglitz J (1980) Industrial structure and the nature of innovative activity. Economic Journal, 90, 266–293. Debreu G (1959) Theory of Value. An Axiomatic Analysis of Economic Equilibrium. Yale University Press. New Haven, CT. Debreu G (1970) Economies with a finite set of equilibria. Econometrica, 38, 387–392. Demsetz H (1967) Toward a theory of property rights. American Economic Review, 57, 347–359. Diamond P (1965) National debt in a neoclassical growth model. American Economic Review, 55, 1126–1150. Dierker E (1974) Topological Methods in Walrasian Economics. Springer-Verlag. Berlin. Dixit A and Stiglitz J (1977) Monopolistic competition and optimum product diversity. American Economic Review, 67, 297–308. Dopfer K (ed) (2005) The Evolutionary Foundations of Economics. Cambridge University Press. Cambridge. Dopfer K and Potts J (2008) The General Theory of Economic Evolution. Routledge. London. Dorfman R, Samuelson P and Solow R (1958) Linear Programming and Economic Analysis. McGrawHill. New York, NY. Dosi G (1982) Technological paradigms and technological trajectories. Research Policy, 11, 147–162. Dosi G (1984) Technical Change and Industrial Transformation. Macmillan. London. Dosi G (1988) Sources, procedures and microeconomic effects of innovation. Journal of Economic Literature, 26, 1120–1171. Dosi G (1995) Hierarchies, markets and power. Industrial and Corporate Change, 4, 1–19. Dosi G (1997) Opportunities, incentives and the collective patterns of technical change. Economic Journal, 107, 1530–1547.

62

The foundational evolutionary traverse of Nelson and Winter Dosi G (2000) Innovation, Organization and Economic Dynamics. Edward Elgar. Cheltenham. Dosi G (2023) The Foundations of Complex Evolving Economies. Oxford University Press. New York, NY. Dosi G, Ermoliev YM and Kaniovsky Y (1994) Generalized urn schemes and technological dynamics. Journal of Mathematical Economics, 23, 1–19. Dosi G, Fagiolo G, Napoletano M and Roventini A (2013) Income distribution, credit and fiscal policies in an agent-based Keynesian model. Journal of Economic Dynamics and Control, 37, 1598–1625. Dosi G, Fagiolo G and Roventini A (2010) Schumpeter meeting Keynes: A policy-friendly model of endogenous growth and business cycles. Journal of Economic Dynamics and Control, 34, 1748–1767. Dosi G, Freeman C, Nelson R, Silverberg G and Soete L (eds) (1988) Technical Change and Economic Theory. Pinter. London. Dosi G and Grazzi M (2010) On the nature of technologies. Cambridge Journal of Economics, 34, 173–184. Dosi G, Marengo L and Fagiolo G (2005) Learning in evolutionary environments. In Dopfer K (ed) The Evolutionary Foundations of Economics. Cambridge University Press. Cambridge. Dosi G, Marengo L and Nuvolari A (2020) Institutions and economic change: Some notes on selforganization, power and learning in human organizations. Eurasian Business Review, 10, 1–22. Dosi G, Marsili O, Orsenigo L and Salvatore R (1995) Learning, market selection and the evolution of industrial structures. Small Business Economics, 39, 1043–1067. Dosi G and Nelson R (2010) Technical change and industrial dynamics as an evolutionary process. In Hall B and Rosenberg N (eds) Handbook of the Economics of Innovation. Elsevier. Amsterdam. Dosi G, Nelson RR and Winter SG (2000) The Nature and Dynamics of Organizational Capabilities. Oxford University Press. New York, NY. Dosi G and Nuvolari A (2020) Introduction: Chris Freeman’s history, coevolution and economic growth. An affectionate reappraisal. Industrial and Corporate Change, 29, 1021–1034. Dosi G, Pereira M and Virgilito E (2017) The footprint of evolutionary processes of learning and selection upon the statistical properties of industrial dynamics. Industrial and Corporate Change, 26, 187–210. Dosi G and Roventini A (2019) More is different and complex. The case for agent-based macroeconomic. Journal of Evolutionary Economics, 29, 1–37. Eggertsson G and Krugman P (2012) Debt, deleveraging and the liquidity trap: A Fisher-Minsky-Koo approach. Quarterly Journal of Economics, 127, 1469–1513. Fatas-Villafranca F, Fernandez C and Vazquez F (2019) Consumer social learning and industrial dynamics. Economics of Innovation and New Technology, 28, 119–141. Fatas-Villafranca F, Jarne G and Sanchez J (2009) Industrial leadership in science-based industries. Journal of Economic Behavior and Organization, 72, 390–407. Fatas-Villafranca F, Jarne G and Sanchez J (2012) Innovation, cycles and growth. Journal of Evolutionary Economics, 22, 207–233. Fatas-Villafranca F, Sanchez J and Jarne G (2008) Modeling the co-evolution of national industries and institutions. Industrial and Corporate Change, 17, 65–108. Fatas-Villafranca F, Saura D and Vazquez F (2007) Emulation, prevention and social interaction in consumption dynamics. Metroeconomica, 58, 582–608. Fatas-Villafranca F, Saura D and Vazquez FJ (2011) A dynamic model of public opinion formation. Journal of Public Economic Theory, 13, 417–441. Feller W (1957) An Introduction of Probability Theory and Its Applications. Wiley. New York, NY. Fisher RA (1930) The Genetic Theory of Natural Selection. Clarendon Press. Oxford. Foster J and Metcalfe JS (eds) (2001) Frontiers of Evolutionary Economics. Edward Elgar. Cheltenham. Freeman C (1982) The Economics of Industrial Innovation. (2nd ed). Pinter. London. Freeman C (1987) Technology Policy and Economic Performance. Pinter. London. Freeman C (ed) ( 1990) The Economics of Innovation. Edward Elgar. Aldershot. Freeman C (2019) History, co-evolution and economic growth. Industrial and Corporate Change, 28, 1–44. Friedman M (1953) Essays in Positive Economics. University of Chicago Press. Chicago, IL. Friedman M (1957) A Theory of the Consumption Function. Princeton University Press. Princeton, NJ. Friedman M (1968) The role of monetary policy. American Economic Review, 58, 1–17. Friedman M (1970) A theoretical framework for monetary analysis. Journal of Political Economy, 78, 193–238.

63

Isabel Almudi and Francisco Fatas-Villafranca Frisch R (1933) Propagation and impulse problems in dynamic economics. In Economic Essays in Honor of Gustav Cassel. Allen and Unwin. London. Fudenberg D and Tirole J (2000) Game Theory. (7th ed). The MIT Press. Cambridge, MA. Fukuyama F (1992) The End of History and the Last Man. Free Press. New York, NY. Gale D (1955) The law of supply and demand. Mathematica Scandinavica, 3, 155–169. Gale D (1956) The closed linear model of production. In Kuhn H and Tucker A (eds) Linear Inequalities and Related Systems. Princeton University Press. Princeton, NJ. Gale D (1960) The Theory of Linear Economic Models. University of Chicago Press. Chicago, IL. Gallegati M, Palestrini A and Russo A (2017) Agent-Based Economics. Elsevier. London. Grossman G and Helpman E (1991) Quality ladders in the theory of growth. Review of Economic Studies, 58, 43–61. Haavelmo T (1943) The statistical implications of a system of simultaneous equations. Econometrica, 11, 1–12. Hall RE (1990) The Rational Consumer. MIT Press. Cambridge, MA. Hanusch H and Pyka A (eds) (2007) Elgar Companion to Neo-Schumpeterian Economics. Edward Elgar. Cheltenham. Hart S and Mas-Colell A (2013) Simple Adaptive Strategies. World Scientific. London. Hayek FA (1937) Economics and knowledge. Economica, 4, 33–54. Hayek FA (1945) The use of knowledge in society. American Economic Review, 35, 519–530. Hayek FA (1967) Studies in Philosophy, Politics and Economics. University of Chicago Press. Chicago, IL. Hayek FA (1988) The Fatal Conceit. University of Chicago Press. Chicago, IL. Hicks JR (1937) Mr. Keynes and the “classics”: A suggested interpretation. Econometrica, 5, 147–159. Hicks JR (1939) Value and Capital. Clarendon Press. Oxford. Hodgson GM (1999) Evolution and Institutions. Edward Elgar. Cheltenham. Hodgson GM (2001) How Economics Forgot History. Routledge. London. Hodgson GM (2015) Conceptualizing Capitalism. University of Chicago Press. Chicago, IL. Hodgson GM (2019) Evolutionary Economics: Its Nature and Future. Cambridge University Press. Cambridge. Hodgson GM and Knudsen T (2010) Darwin’s Conjecture. Chicago University Press. Chicago, IL. Hofbauer J and Sigmund K (1998) Evolutionary Games and Population Dynamics. Cambridge University Press. Cambridge. Jackson MO (2008) Social and Economic Networks. Princeton University Press. Princeton, NJ. Kahneman D (2003) Maps of bounded rationality: Psychology for behavioral economics. American Economic Review, 93, 1449–1475. Keynes JM (1936) The General Theory of Employment, Interest and Money. Macmillan. London. Keynes JM (1937) The general theory of employment. Quarterly Journal of Economics, 51, 209–223. Kirzner IM (1992) The Meaning of Market Process. Routledge. London. Klein B (1977) Dynamic Competition. Harvard University Press. Cambridge, MA. Klein L and Goldberger A (1955) Economic Fluctuations in the United States, 1929–1952. North Holland. Amsterdam. Knight F (1921) Risk, Uncertainty and Profit. Houghton Mifflin. Boston, MA. Knight F (1951) The Economic Organization. Routledge. London. Koopmans T (1936) Linear Regression Analysis of Economic Time Series. F Bohn. New York, NY. Koopmans T (ed) (1950) Statistical Inference in Dynamic Economic Models. Wiley. New York, NY. Koopmans T (ed) (1951) Activity Analysis of Production and Allocation. Wiley. New York, NY. Koopmans T (1957) Three Essays on the State of Economic Science. McGraw Hill. New York, NY. Kydland F and Prescott E (1982) Time to build and aggregate fluctuations. Econometrica, 50, 1345–1371. Kuznets S (1929) Secular Movements in Prices and Production. Houghton Mifflin. Boston, MA. Ljungqvist L and Sargent T (2018) Recursive Macroeconomic Theory. (4th ed) MIT Press. Cambridge, MA. Loasby B (1999) Knowledge, Institutions and Evolution in Economics. Routledge. London. Lucas RE (1972) Expectations and the neutrality of money. Journal of Economic Theory, 4, 103–124. Lucas RE (1983) Studies in Business Cycle Theory. MIT Press. Cambridge, MA. Lucas RE (1988) On the mechanics of economic development. Journal of Monetary Economics, 22, 3–42.

64

The foundational evolutionary traverse of Nelson and Winter Lucas RE and Prescott E (1971) Investment under uncertainty. Econometrica, 39, 659–681. Lundvall BA (1992) National Systems of Innovation. Pinter. London. Malerba F (1985) The Semi-Conductor Business: The Economics of Rapid Growth and Decline. University of Wisconsin Press. Madison, WI. Malerba F (2002) Sectoral systems of innovation and production. Research Policy, 31, 247–264. Malerba F and Nelson (eds) (2013) Economic Development as a Learning Process. Edward Elgar. Cheltenham. Malerba F, Nelson R, Orsenigo L and Winter S (2016) Innovation and the Evolution of Industries: History-Friendly models. Cambridge University Press. Cambridge, NY. Marschak J and Radner R (1972) Economic Theory of Teams. Yale University Press. New Haven, CT. Mas-Colell A (1985) The Theory of General Economic Equilibrium: A differentiable approach. Cambridge University Press, Cambridge. Maynard-Smith J (1982) Evolution and the Theory of Games. Cambridge University Press. Cambridge. Mazzucato M (2013) The Entrepreneurial State. Anthem Press. London. Metcalfe JS (1994) Competition, Fisher’s principle and increasing returns in the selection process. Journal of Evolutionary Economics, 4, 327–346. Metcalfe JS (1998) Evolutionary Economics and Creative Destruction. Routledge. London. Metcalfe JS (2001) Institutions and progress. Industrial and Corporate Change, 10, 561–586. Metcalfe JS (2010) Technology and economic theory. Cambridge Journal of Economics, 34, 153–171. Metcalfe JS, Ramlogan R and Foster J (2006) Adaptive economic growth. Cambridge Journal of Economics, 30, 7–32. Miao J (2022) Economic Dynamics in Discrete Time. (2nd ed). MIT Press. Cambridge, MA. Mises L (1949) Human Action. Yale University Press. New Haven, CT. Mowery D and Nelson R (eds) (1999) The Sources of Industrial Leadership. Cambridge University Press. New York, NY. Muñoz FF, Encinar MI and Cañibano C (2011) On the role of intentionality in evolutionary economic change. Structural Change and Economic Dynamics, 22, 193–203. Murmann JP (2003) Knowledge and Competitive Advantage: The Coevolution of Firms, Technology and National Institutions. Cambridge University Press. Cambridge, NY. Muth JF (1961) Rational expectations and the theory of price movements. Econometrica, 29, 315–335. Nelson RR (1956) A theory of the low level equilibrium trap in developing countries. American Economic Review, 46, 894–908. Nelson RR (1959) The simple economics of basic scientific research. A theoretical analysis. Journal of Political Economy, 67, 297–306. Nelson RR (ed) (1962) The Rate and Direction of Inventive Activity. Princeton University Press. Princeton, NJ. Nelson RR (1964) Aggregate production functions and medium-range growth projections. American Economic Review, 54, 575–606. Nelson RR (1968) A diffusion model of international productivity differences in manufacturing industry. American Economic Review, 58, 1219–1248. Nelson RR (1977) The Moon and the Ghetto. Norton. New York, NY. Nelson RR (1980) Production sets, technological knowledge and R&D: Fragile and overworked constructs for the analysis of productivity growth? American Economic Review, 70, 62–67. Nelson RR (1981) Research on productivity growth and differences. Journal of Economic Literature, 19, 1029–1064. Nelson RR (1982) The role of knowledge in R&D efficiency. Quarterly Journal of Economics, 97, 453–470. Nelson RR (ed) (1993) National Innovation Systems. Oxford University Press. New York, NY. Nelson RR (1996) The Sources of Economic Growth. Harvard University Press. Cambridge, MA. Nelson RR (1998) The agenda for growth theory: A different point of view. Cambridge Journal of Economics, 22, 497–520. Nelson RR (2005) Technology, Institutions and Economic Growth. Harvard University Press. Cambridge, MA. Nelson RR (2008) What enables rapid economic progress: What are the needed institutions? Research Policy, 37, 1–11. Nelson RR (2013) Demand, supply and their interaction in markets as seen from the perspective of evolutionary economic theory. Journal of Evolutionary Economics, 23, 17–38.

65

Isabel Almudi and Francisco Fatas-Villafranca Nelson RR (2016) The behavior and cognition of economic actors in evolutionary economics. Journal of Evolutionary Economics, 26, 737–751. Nelson RR (2018) Modern Evolutionary Economics. An Overview. Cambridge University Press. New York, NY. Nelson RR (2022) The economic system question revisited. Industrial and Corporate Change, 31, 591–609. Nelson RR and Pack H (1999) The Asian miracle and modern growth theory. Economic Journal, 109, 416–436. Nelson RR and Phelps ES (1966) Investment in humans, technological diffusion and economic growth. American Economic Review, 56, 69–75. Nelson RR and Winter SG (1964) A case study in the economics of information and coordination: The weather forecasting system. Quarterly Journal of Economics, 78, 420–441. Nelson RR and Winter SG (1973) Toward an evolutionary theory of economic capabilities. American Economic Review, 63, 440–449. Nelson RR and Winter SG (1974) Neoclassical vs Evolutionary theories of economic growth: Critique and prospectus. Economic Journal, 84, 886–905. Nelson RR and Winter SG (1977) In search of a useful theory of innovation. Research Policy, 6, 36–76. Nelson RR and Winter SG (1978) Forces generating and limiting concentration under Schumpeterian competition. The Bell Journal of Economics, 9, 524–548. Nelson RR and Winter SG (1982) An Evolutionary Theory of Economic Change. The Belknap Press of Harvard University Press. Cambridge, MA. Nelson RR and Winter SG (2002) Evolutionary theorizing in economics. Journal of Economic Perspectives, 16, 23–46. North DC (1991) Institutions, Institutional Change and Economic Performance. Cambridge University Press. Cambridge, NY. North DC (2003) Understanding the Process of Economic Change. Princeton University Press. Princeton, NJ. Pasinetti L (1981) Structural Change and Economic Growth. Cambridge University Press. Cambridge. Pavitt K (1984) Sectoral patterns of technical change: Towards a taxonomy and a theory. Research Policy, 13, 343–373. Phelps ES (1970) Microeconomic Foundations of Employment and Inflation Theory. Norton. New York, NY. Phelps ES and Winter SG (1970) Optimal price policy under atomistic competition. In Phelps E (ed) Microeconomic Foundations of Employment and Inflation Theory. Norton. New York, NY. Polanyi M (1962) Personal Knowledge: Towards a Post-Critical Philosophy. Harper. New York, NY. Polanyi M (1967) The Tacit Dimension. Doubleday-Anchor. New York, NY. Popper K (1945) The Open Society and Its Enemies. Routledge. London. Potts J (2000) The New Evolutionary Microeconomics. Edward Elgar. Cheltenham. Price GR (1972) Extension of covariance selection mathematics. Annals of Human Genetics, 35, 485–490. Pyka A (2017) Dedicated innovation systems to support the transformation towards sustainability. Journal of Open Innovation: Technology, Market, Complexity, 3, 27. Radner R (1961) Paths of economic growth that are optimal with regard only to final states: A turnpike theorem. Review of Economic Studies, 28, 98–104. Rawls J (1971) A Theory of Justice. Harvard University Press. Cambridge, MA. Robbins L (1932) An Essay on the Nature and Significance of Economic Science. Macmillan. London. Romer D (2012) Advanced Macroeconomics. McGraw Hill. New York, NY. Romer PM (1990) Endogenous technological change. Journal of Political Economy, 98, 71–102. Rosenberg N (1976) Perspectives on Technology. Cambridge University Press. Cambridge. Rosenberg N (1982) Inside the Black Box. Cambridge University Press. Cambridge. Rothblum U and Winter SG (1985) Asymptotic behavior of market shares for a stochastic growth model. Journal of Economic Theory, 36, 352–366. Samuelson PA (1947) Foundations of Economic Analysis. Harvard University Press. Cambridge, MA. Samuelson PA (1958) An exact consumption-loan model of interest with or without the social contrivance of money. Journal of Political Economy, 66, 467–482. Samuelson L (1997) Evolutionary Games and Equilibrium Selection. MIT Press. Cambridge, MA. Sandholm WH (2010) Population Games and Evolutionary Dynamics. MIT Press. Cambridge, MA.

66

The foundational evolutionary traverse of Nelson and Winter Sargent TJ (1987) Dynamic Macroeconomic Theory. Harvard University Press. Cambridge, MA. Saviotti PP (1996) Technological Evolution and Economic Variety. Edward Elgar. Cheltenham. Saviotti PP and Pyka A (2004) Economic development by the creation of new sectors. Journal of Evolutionary Economics, 14, 1–35. Saviotti PP and Pyka A (2013) The coevolution of innovation, demand and growth. Economics of Innovation and New Technology, 22, 461–482. Schumpeter JA (1934) The Theory of Economic Development. Harvard University Press. Cambridge, MA (1st ed. German 1911). Schumpeter JA (1942) Capitalism, Socialism and Democracy. Harper Row. New York, NY. Schuster P and Sigmund K (1983) Replicator dynamics. Journal of Theoretical Biology, 100, 535–538. Sen A (1970) Collective Choice and Social Welfare. Holden-Day. San Francisco, CA. Shackle GLS (1979) Imagination and the Nature of Choice. Edinburgh University Press. Shiozawa Y, Morioka M and Taniguchi K (2019) Microfoundations of Evolutionary Economics. Springer Japan. Tokyo. Silverberg G, Dosi G and Orsenigo L (1988) Innovation, diversity and diffusion: A self-organization model. Economic Journal, 98, 1032–1054. Silverberg G and Soete L (1994) The Economics of Growth and Technical Change. Edward Elgar. Cheltenham. Silverberg G and Verspagen B (1994) Collective learning, innovation and growth in a boundedly rational evolutionary world. Journal of Evolutionary Economics, 4, 207–226. Silverberg G and Verspagen B (2005) A percolation model of innovation in complex technology spaces. Journal of Economic Dynamics and Control, 29, 225–244. Simon H (1947) Administrative Behaviour. MacMillan. New York, NY. Simon H (1957) Models of Man. Wiley. New York, NY. Simon H (1983) Reason in Human Affairs. Stanford University Press. Stanford, CA. Solow R (1956) A contribution to the theory of economic growth. Quarterly Journal of Economics, 70, 65–94. Solow R (1957) Technical change and the aggregate production function. Review of Economics and Statistics, 39, 312–320. Stokey NL, Lucas RE and Prescott EC (1989) Recursive Methods in Economic Dynamics. Harvard University Press. Cambridge, MA. Teece DJ, Pisano G and Shuen A (1997) Dynamic capabilities and strategic management. Strategic Management Journal, 18, 509–533. Theil H (1978) Introduction to Econometrics. Prentice-Hall. Englewood Cliffs, NJ. Tinbergen J (1939) Business Cycles in the United States, 1919–1932. League of Nations. Geneva. Tirole J (1988) The Theory of Industrial Organization. MIT Press. Cambridge, MA. Tobin J (1955) A dynamic aggregative model. Journal of Political Economy, 63, 103–115. Tobin J (1965) Money and economic growth. Econometrica, 33, 671–684. Tobin J (1969) A general equilibrium approach to monetary theory. Journal of Money, Credit and Banking, 1, 15–29. Valente M (2012) Evolutionary demand: A model of boundedly rational consumers. Journal of Evolutionary Economics, 22, 1029–1080. Valente M and Andersen E (2002) A hands-on approach to evolutionary simulation: Nelson and Winter models in LSD. The e-Journal of Evolutionary Modeling and Economic Dynamics, 1003 ( http://beagle.u-bordeaux4.fr/jemed/1003/1003.pdf) Valente M, Lorentz A, Ciarli T and Savona M (2015) The effect of demand-driven structural transformations on growth and technological change. Journal of Evolutionary Economics, 26, 219–246. Varian HR (1980) Microeconomic Analysis. Norton. New York, NY. Veblen T (1898) Why is economics not an evolutionary science? Quarterly Journal of Economics, 12, 373–397. Verspagen B (1991) A new empirical approach to catching-up or falling behind. Structural Change and Economic Dynamics, 2, 359–380. Von Neumann J (1945) A model of general economic equilibrium. Review of Economic Studies, 13, 1–9. Von Neumann J and Morgenstern O (1944) Theory of Games and Economic Behavior. Princeton University Press: Princeton. Weibull J (1995) Evolutionary Game Theory. MIT Press. Cambridge, MA.

67

Isabel Almudi and Francisco Fatas-Villafranca Williamson OE (1975) Markets and Hierarchies. Free Press. New York, NY. Williamson OE (1985) The Economic Institutions of Capitalism. Free Press. New York, NY. Williamson OE and Winter SG (1991) The Nature of the Firm: Origins, Evolution, Development. Oxford University Press. New York, NY. Wilson DS and Kirman A (eds) (2016) Complexity and Evolution. The MIT Press. Cambridge, MA. Winter SG (1964) Economic “natural selection” and the theory of the firm. Yale Economic Essays, 4, 225–272. Winter SG (1965) Some properties of the closed linear model of production. International Economic Review, 6, 199–210. Winter SG (1967) The norm of a closed technology and the straight-down-the-Turnpike theorem. Review of Economic Studies, 34, 67–84. Winter SG (1969) A simple remark on the second optimality theorem of welfare economics. Journal of Economic Theory, 1, 99–103. Winter SG (1971) Satisficing, selection and the innovating remnant. Quarterly Journal of Economics, 85, 237–261. Winter SG (1981) Attention allocation and input proportions. Journal of Economic Behaviour and Organization, 2, 31–46. Winter SG (1982) An essay on the theory of production. In Hyman S (ed) Economics and the Worlds Around It. University of Michigan Press. Ann Arbor, MI. Winter SG (1984) Schumpeterian competition in alternative technological regimes. Journal of Economic Behaviour and Organization, 5, 287–320. Winter SG (1986) Adaptive behaviour and economic rationality: Comments on Arrow and Lucas. Journal of Business, 59, 427–434. Winter SG (1992) The Budget Deficit and Long-term Economic Growth. Testimony before the US Congress ( https://www.gao.gov/assets/t-oce-92-1.pdf). Winter SG (1995) Four Rs of profitability: Rents, resources, routines and replication. In Montgomery C (ed) Resource-Based and Evolutionary Theories of the Firm: Towards a Synthesis. Kluwer Academic Publishers. Boston, MA. Winter SG (2003) Understanding dynamic capabilities. Strategic Management Journal, 24, 991–995. Winter SG (2005) Towards an evolutionary theory of production. In Dopfer K (ed) The Evolutionary Foundations of Economics. Cambridge University Press. Cambridge. Winter SG (2006) Toward a Neo-Schumpeterian theory of the firm. Industrial and Corporate Change, 15, 125–141. Winter SG (2008) Scaling heuristics shape technology. Should economic theory take notice? Industrial and Corporate Change, 17, 513–531. Winter SG (2010) Testimony Submitted to the US Congress, House Science and Technology Committee ( www.lem.sssup.it/paper_seminars/budapest/winter.pdf). Winter SG (2014) The future of evolutionary economics: Can we break out of the beachhead? Journal of Institutional Economics, 10, 613–644. Winter SG (2017) Pursuing the evolutionary agenda in economics and management research. Cambridge Journal of Economics, 41, 721–747. Winter SG, Cattani G and Dorsch A (2007) The value of moderate obsession: Insights from a new model of organizational search. Organization Science, 18, 403–419. Winter SG, Dosi G and Kaniovski Y (2003) A baseline model of industry evolution. Journal of Evolutionary Economics, 13, 355–383. Witt U (2001) Learning and the theory of wants. A theory of wants and the growth of demand. Journal of Evolutionary Economics, 11, 23–36. Witt U (2003) The Evolving Economy. Edward Elgar. Cheltenham. Witt U (2009) Novelty and the bounds of un-knowledge in economics. Journal of Economic Methodology, 16, 361–375. Witt U (2014) The future of evolutionary economics: Why the modalities of explanation matter? Journal of Institutional Economics, 10, 645–664. Woodford M (2003) Interest and Prices. Princeton University Press. Princeton, NJ. Zollo M and Winter SG (2002) Deliberate learning and the evolution of dynamic capabilities. Organization Science, 13, 339–351.

68

4 F.A. HAYEK AND EVOLUTIONARY AUSTRIAN ECONOMICS Viktor J. Vanberg

4.1

Carl Menger and his heirs

The label “Austrian School of Economics” comprises a group of scholars who are united less by a shared paradigm than by their common intellectual heritage, a research tradition that goes back to the school’s founder, Carl Menger. Scholars associated with the school consider themselves Carl Menger’s intellectual heirs and subscribe to what are considered the principal ingredients of Menger’s theoretical outlook, his methodological individualism and subjectivism. Yet, there exist significant divisions within the school on how these principles ought to be interpreted, divisions that separate, in particular, the followers of Ludwig von Mises – who consider theirs to be the only “correct Austrian paradigm” (Rothbard 1992: 7) – from those who find more merits in the research program advanced by Friedrich A. Hayek. Taking the Misesian and the Hayekian versions of the “Austrian paradigm” as the principal rivals in today’s Austrian economics, in what follows I shall argue in support of three claims: • Firstly, the claim that the “causal-genetic method,” central to Menger’s work, suggests itself as nucleus of an evolutionary research program. • Secondly, the claim that among Menger’s heirs F.A. Hayek is the one who took up most systematically the evolutionary elements in the founder’s work and made them a central part of his own research efforts. • Thirdly, the claim that, if Austrian economics aspires to be recognized as an empirical social science, it is the Mengerian-Hayekian evolutionary paradigm, not Misesian praxiology, that provides a research program suitable for this purpose.

4.2

Carl Menger’s evolutionary social theory

As Naomi Beck and Ulrich Witt (2019: 205) observe in their “Austrian Economics and the Evolutionary Paradigm,” Menger does not explicitly invoke Darwin’s theory, yet his approach to social phenomena clearly aims at achieving what Darwin accomplished for

DOI: 10.4324/9780429398971-6

69

Viktor J. Vanberg

natural phenomena. Just as Darwin succeeded in showing how a causal explanation can be provided for natural phenomena that appear to be the product of purposeful design, Menger sees the major explanatory challenge the social sciences face in showing how social phenomena that seem to be “the result of socially teleological causes” ([1871] 1981: 158 f.) can be explained as “the unintended social result of individually teleological factors” ([1883] 1985: 158), i.e. of individual actions motivated by individual purposes. The “atomistic,” “compositive” or “causal-genetic”1 method that such study requires must, so Menger argues, aim at explaining social phenomena by “reducing them to their elements, to the individual factors of their causation, and by investigating the laws by which the complicated phenomena … are built up from these elements” ([1883] 1985: 159). It is, he emphasizes, the same method that must also be applied in the study of economic phenomena such as “prices of goods, interest rates, ground rents, wages” ([1883] 1985: 147).2 Distinguishing between social phenomena of “organic origin” – “the unintended result of human efforts aimed at essentially individual goals” (ibid.:133) – and social phenomena of “pragmatic origin” – the “result of purposeful activity of humans directed towards their establishment” (ibid.: 132) – Menger recognizes that “for the understanding of social phenomena in their entirety the pragmatic interpretation is, in any case, just as indispensable as the ‘organic’” (ibid.: 135). Yet, explaining the origin and function of purposefully created social phenomena and institutions he does not consider a particularly challenging task. Rather, what he refers to as “perhaps the most noteworthy problem of the social sciences” (ibid.) he sees in providing an answer to the question: How can it be that institutions which serve the common welfare and are extremely significant for its development come into being without a common will directed toward establishing them (ibid.: 146). The explanatory challenge an “organic understanding” of social phenomena must meet is, as Menger notes, to understand them theoretically as resulting from “the ultimate and general cause of all economic activity, the endeavor of men to … better their economic positions” ([1871] 1981: 192). The task is to provide a causal-genetic account of the sequence of events by “which the more complex economic phenomena evolve from their elements according to definite principles” (ibid.: 46 f.). A paradigm example of such process-analysis Menger has provided with his famous theory of the “nature and origin of money” ([1871] 1981: 257–262; [1883] 1985: 152–155; 1892), taking as his starting point a barter economy in which trade is limited to situations in which economizing individuals have goods in their possession that have a smaller use value to them than goods in the possession of other economizing individuals who value the same goods in reverse fashion ([1871] 1981: 258). Facing the difficulty of finding a matching counterpart, Menger argues, an economizing individual will learn that he can get nearer to his final end if instead of trading only with someone who offers goods that have use value to him, he accepts in exchange goods that “have greater marketability than his own commodity” (ibid.: 259). As more saleable goods are preferable to less saleable ones, trades will over time converge toward a common and generally accepted medium of exchange or “‘money’ in the broadest sense of the word” ([1883] 1985: 152). 70

F.A. Hayek and evolutionary Austrian economics

The genetic-causal method of explanation that he exemplifies in detail for the origin of money can, so Menger suggests, be equally applied to other social structures and institutions, such as law, language, morals, the origin of communities or of states (ibid.: 130, 147, 155 f.). Whether, in fact, as Menger supposes, the kind of evolutionary explanation that his theory of the origin of money exemplifies is applicable to social institutions in general, is an issue that will be addressed below in the context of Hayek’s theory of cultural evolution.

4.3

F.A. Hayek’s Darwinian theory of cultural evolution

In his own contribution to a theory of cultural evolution Hayek expressly, and repeatedly, credits Carl Menger with having revived the explanatory approach to social phenomena that the 18th-century “Scottish founders” of political economy, David Hume, Adam Ferguson, and Adam Smith had developed (Hayek 1984: 319). As “Darwinians before Darwin” ([1967] 1978: 264f.) they advanced an evolutionary social theory showing, as Hayek (1960: 59) puts it: that an evident order which was not the product of a designing human intelligence need not therefore be ascribed to the design of a higher, supernatural intelligence, but that there was a third possibility – the emergence of order as the result of adaptive evolution. The evolutionary tradition that Menger revived was, though, not the component of his work that his followers adopted as their research program.3 To the extent that this tradition has found its place in today’s Austrian economics it is due to the work of F.A. Hayek who, “yet another eighty years later” (Hayek [1967] 2014: 297), followed up on Menger’s project. While the Scottish moral philosophers deserve to be called “Darwinians before Darwin,” it is, Hayek notes, the significant contribution of Darwin’s general theory of evolution to have explicated the basic principles that supply us “with a key to a general understanding of the formation of order in life, mind and interpersonal relations” (Hayek 1988: 144). As he puts it: The basic proposition … is that a mechanism of reduplication with transmittable variations and competitive selection of those which prove to have a better chance of survival will in the course of time produce a great variety of structures adapted to continuous adjustment to the environment and to each other ([1964] 2014: 267). With his formula of the “twin ideas of evolution and spontaneous order” Hayek seeks to capture the essence of a theoretical outlook that characterizes sciences dealing with “such ‘more highly organized’ or essentially complex phenomena as we encounter in the realms of life, mind and society” (1973: 41),4 phenomena in the study of which our explanatory efforts face “inevitable limitations” ([1973] 1978: 278f.). The social sciences, as the “life sciences” in general, face these limitations, Hayek argues, not because of the impossibility to identify the general principles that work in bringing the phenomena they study about. It is rather the impossibility of ascertaining all circumstances of their origin (1973: 41) that causes the “peculiar difficulties” (ibid.) the sciences of life, mind, and society face in their explanatory efforts. And these difficulties, Hayek posits, are inherent in the evolutionary nature of the phenomena they study, in the fact that the processes by which they are brought about are 71

Viktor J. Vanberg

learning- and knowledge-creating processes. As Menger had emphasized in his theory of the emergence of money, while people may be assumed to be “regularly governed by their individual interests” ([1883] 1985: 64), they are not equipped with perfect knowledge of what are effective ways to advance their interests, but learn over time from their own experience and the example of others (ibid.: 261). Accordingly, the fact that their actions reflect the experience individuals have accumulated in their learning history has to be taken into account in efforts to explain their behavior. Being counted, along with Karl R. Popper and Donald Campbell, as one of the founders of evolutionary epistemology (Vanberg 2017: 38 ff.), Hayek looks at evolutionary processes – whether biological or cultural – as processes of learning or the growth of knowledge and, reversely, interprets all learning as the product of evolutionary processes. Biological evolution is viewed as a process in which, through variation and selection, better solutions for the problem of survival are discovered and retained as genetically coded knowledge; personal learning is interpreted as a process in which individuals, through trial and errorelimination, discover better ways for dealing with the various problems they face, acquiring memory-coded knowledge; and, finally, cultural evolution is looked at as a process in which the problem-solving capacity of social groups grows as a result of their experimenting with different conjectural solutions, generating knowledge that is “stored” in cultural achievements such as ordinary tools, standard practices, social rules and customs (Vanberg 1994). In his theory of cultural evolution Hayek likes to draw an analogy between ordinary tools and social rules or institutions as “social tools,” as devices that serve to solve recurrent problems in individuals’ interactions with each other, in the manner in which ordinary tools serve to solve problems they encounter recurrently in dealing with their physical environment.5 Such analogy tends to suggest that the evolutionary processes from which the two kinds of “tools” result are similar in nature and that their origin and maintenance can, accordingly, be explained in the same manner. There are, however, relevant differences between these two kinds of “tools.” While, in the case of ordinary tools, individuals’ immediate self-interest naturally leads them to opt for tools that prove to be more effective “problem solvers,” the same cannot be generally presumed to be the case for “social tools” as well. That in case of the latter individuals’ self-interest does not always motivate them to follow “problem-solving” rules, Hayek obscures when he notes about social rules: Like all general purpose tools, rules serve because they have become adapted to the solution of recurrent problem situations and thereby help to make the members of the group in which they prevail more effective in the pursuit of their aims (1976: 21). The phrase that rules “help to make the members of the group in which they prevail more effective in the pursuit of their aims” is ambiguous. It may mean that following the rule in question is directly beneficial to the rule-follower, or it may mean that it is beneficial to belong to a group in which the rule is generally followed. In this regard rules or institutions, such as property rights, that help to solve coordination problems, such as the institution of money, differ systematically from those solving social dilemmas. In the case of coordination-rules, it is in each individual’s immediate interest to follow them, once they are established in a group, and individuals benefit from living in such a group. Because of such constellation of interests, coordination-rules are typically self-enforcing. In contrast, while it is surely to an individual’s advantage to live in a group in which rules solving social dilemmas, e.g. respecting property rights, are generally adhered to, this fact per se does not produce an 72

F.A. Hayek and evolutionary Austrian economics

immediate interest for individuals to follow them. In other words, such rules are not selfenforcing. To secure compliance with them, incentives are needed that go beyond the benefits derived from others’ following them, a fact Hayek recognizes when he distinguishes between rules which individuals “will follow spontaneously” and those “which they have to be made to obey, since, although it would be in the interest of each to disregard them, the overall order on which the success of their actions depends will arise only if these rules are generally followed” (1973: 45). Realizing that rules of the latter kind cannot be presumed to “evolve” in the same manner as rules which individuals “will follow spontaneously,” Hayek invokes group selection as the mechanism that he supposes to work in these cases. Cultural evolution, so he argues, proceeds simultaneously at two levels: It involves competition between organized and unorganized groups, no less than competition between individuals … The endeavor to achieve certain results by cooperation and organization is as much part of competition as individual efforts. Successful group relations also prove their effectiveness in competition among groups organized in different ways (1960: 37). Though competition between groups is surely as much a fact as competition between individuals, the question remains what one may conclude from this fact regarding the emergence and maintenance of social-dilemma norms. When Hayek argues that such rules “have evolved because the groups who practiced them were more successful and displaced others” (1973: 18), he seems to presume that the benefit rules provide to groups adopting them explains their presence. The problem with such presumption is that it fails to answer the question of how the systematic conflict may be resolved that in the case of socialdilemma rules inevitably exists between the forces of within-group competition between individuals and the forces that drive the competition between groups. To the extent that complying with social-dilemma rules comes with a sacrifice, rule-following group-members are at a disadvantage compared to those who enjoy the benefits created by the rule-followers but spare themselves the sacrifices their own compliance would require – if no compensating incentives are provided within the group. And the ability of groups to create and maintain an effective incentive-mechanism is a cultural achievement that remains always precarious, requires permanent efforts for its maintenance, and that, once acquired, can get lost again where such efforts wane.

4.4

Ludwig von Mises’ praxiological, aprioristic subjectivism

Apart from the various, relatively minor theoretical divisions among Menger’s heirs, the most fundamental divide within the Austrian School is surely between the evolutionary research program represented by F.A. Hayek’s work and the praxeological paradigm that Ludwig von Mises founded. Both research programs claim to be firmly rooted in “the systematic subjectivism and individualism” (Hayek [1952] 2010: 101) which are generally considered the characterizing mark of the theoretical tradition that Menger founded. They categorically differ, however, in how they interpret the theoretical and methodological conclusions that these principles, specifically the subjectivism-part, command. The relevant contrast is between, on Mises’ side, a praxeological, aprioristic subjectivism and, on Hayek’s side, an evolutionary, naturalistic subjectivism.6

73

Viktor J. Vanberg

The fact that “economics starts from acting man’s subjective valuations and the action that is governed by his valuations” (ibid.: 170), so Mises asserts, separates it – and the sciences of human action in general – categorically from the natural sciences. As he puts it: “Reason and experience show us two separate realms: the external world of physical and physiological events and the internal world of thought, feeling, and purposeful action. No bridge connects – as far as we can see today – these two spheres” ([1944] 1990: 25 f.). As for the domain of praxeology he states, that it “deals with purposive behavior, i.e. human action,” in contrast to “a reactive response to stimuli on the part of the bodily instincts, which cannot be controlled by volition” (ibid.: 23). More specifically: It deals with the choosing as such, with the categorical elements of choice and action … , with the pure elements of setting aims and applying means. (ibid.: 21) That its subject is “action as such” means, as Mises explains, that praxeology “does not enter into a discussion of the motives determining the choice” (ibid.: 20). As he notes, “what a man chooses” and the “motives and springs of action are without concern for the praxeological investigation” (ibid.: 21). And for the “statements and propositions” praxeology pronounces Mises (1949: 32) claims that they are not only “like those of logic and mathematics, a priori,” but also refer “with the full rigidity of their apodictic certainty and incontestability to the reality of action as it appears in life and history” (ibid. 39).7 The task of inquiring into the theories that guide and the ends that motivate the actions of concrete persons, Mises expressly assigns to history as the, in this scheme, second branch of the sciences of human action. As he notes: It is the scope of history and not of praxeology to investigate what ends people aim at and what means they apply for the realization of their plans. ([1944] 1990: 24) In examining “the individual and unique conditions” of the cases it deals with, history employs the “specific method” of “understanding” (Mises ([1942] 1990: 12). It establishes, so he notes, “the fact that an individual or a group of individuals have engaged in a definite action emanating from definite judgements of value and choices and aiming at definite ends” ([1944] 1990: 27). In Mises’ account, its specific method of “understanding” sets history categorically apart from the natural sciences, in the same way as its apriorism does with praxeology. A methodological dualism separates, in his view, both branches of the social sciences, dealing with the internal world of thought, feeling and purposive actions, from the natural sciences dealing with the external world of physical and physiological events ([1944] 1990: 25). The claim that a methodological dualism categorically separates the social sciences from the natural sciences Hayek challenges with his evolutionary subjectivism, contending that a naturalistic, causal account of purposive human action can, at least in principle, be given.

4.5

F.A. Hayeks’s evolutionary, naturalistic subjectivism

About “Economics and Knowledge,” his 1936 presidential address to the London Economic Club, Hayek has noted in retrospect that it marked an important shift in his research orientation ([1965] 2014: 49f.), describing it as “an attempt to persuade Mises himself that when he asserted that the market theory was a priori, he was wrong” (1994: 72). 74

F.A. Hayek and evolutionary Austrian economics

At the very beginning of the esssay he states as his “main contention … that the empirical element in economic theory – the only part which is concerned not merely with implications but with causes and effects … – consists of propositions about the acquisition of knowledge” ([1937] 2014a: 57). His purpose is, as he notes, “to restore to its rightful place the investigation of causal processes” (ibid.: 59), in contrast to “the tautological transformations of the Pure Logic of Choice” (ibid.: 62). And about the inquiry into these “causal processes” he states: [T)he assumptions or hypotheses, which we have to introduce when we want to explain the social processes, concern the relation of the thought of an individual to the outside world, the question to what extent and how his knowledge corresponds to the external facts. And the hypotheses must necessarily run in terms of assertion about causal connections, about how experience creates knowledge. (ibid.: 69) With his naturalistic, evolutionary subjectivism Hayek provides an answer to a challenge Mises ([1944] 1990: 26) poses when he states: Identical external events result sometimes in different human responses, and different external events produce sometimes the same human response. We do not know why. In contrast to Mises’ claim that “the mental activities of men that determine their actions, … the reactions of the mind to the conditions of the individual’s environment, … cannot be perceived by the methods of the natural sciences” ([1962] 2006: 43), Hayek insists that, even if it may not be possible to “perceive” them “by the methods of the natural sciences,” these “mental activities” can be studied in terms of the empiricist, causal methodology of the natural sciences. In his account, individuals’ subjective valuation and theories are intervening variables, the link between the environmental conditions they face and the behavior they exhibit, variables that can be accounted for by “empirical propositions … about how people will learn” (Hayek 1979: 57). In Hayek’s approach to social theory, three levels of evolutionary learning play a role in explaining human behavior: the “knowledge” of relevant environmental contingencies the human species accumulated over its evolutionary history, the knowledge individuals accumulate in the course of their personal history, and the experience social groups accumulate in the process of cultural evolution. In the first case, the relevant knowledge is stored in the organisms’ genetic endowment; in the second case, it is coded in individuals’ memory; and in the third case, it is embodied in cultural traditions. In each case, Hayek posits, the relevant knowledge consists in conjectures or theories about contingencies in the environment, contingencies that are of relevance for solving problems the organism, the individual, or the group face.8 As he puts it: “all we know about the world is of the nature of theories and all ‘experience’ can do is to change these theories” (2017: 259). The theory-based expectations evolutionary learning generates – Hayek refers to them as dispositions – allow individuals to anticipate the likely consequences of potential alternative courses of actions and to behave in ways that promise to be most serviceable to the goals they pursue. Accordingly, purposive or goal-directed behavior can, in Hayek’s view, explained as being evoked by dispositions that themselves can be explained as the result of evolutionary learning, their genetically coded dispositions reflecting the evolutionary history of the species, their memory-coded dispositions reflecting their personal learning history. 75

Viktor J. Vanberg

In other words, individuals’ dispositions are the proximate causes, their evolutionary and learning histories the ultimate causes of their behavior.

4.6

Conclusion

At the end of their study on “Austrian Economics and the Evolutionary Paradigm,” Beck and Witt note that, in regard to “the future development of Austrian Economics” (2019: 205), its adherents face a choice between following “the Misesian canon of concepts and political tenets” (ibid.: 220) or following the lead of Hayek’s evolutionary and naturalistic approach. The choice should not be difficult to make if, as is generally asserted, continuing and further developing the research program that Carl Menger initiated gives the Austrian school its identity. There is little evidence to be found in Menger’s work that would justify the claim that the Misesian paradigm represents the only “correct Austrian paradigm” (Rothbard 1992: 7), or that would lend support to the assertion “that Carl Menger’s ideas and contributions were the starting point for the development of Ludwig von Mises’ epistemology and methodology” (Sanz Bas et al. 2020: 423). One may certainly be able to quote statements from Menger’s work that seem to support such claims, yet a closer look at the respective context and the general thrust of his work suggest the opposite. When, for instance, Menger asserts that it “would be improper … to attempt a natural-scientific orientation of our science” ([1871] 1981: 47), this is, quite apparently, not meant as an argument for methodological dualism. It is meant as an argument against “attempts to carry over the peculiarities of the natural-scientific method of investigation uncritically into economics” (ibid.: 47), and as a reminder that the respective fields of knowledge are dealing with “laws peculiar to each field” (ibid.). About the “method of research” that he advocates, and about the “laws peculiar to each field,” Menger states: This method of research, attaining universal acceptance in the natural sciences, … came mistakenly to be called the natural scientific method. It is, in reality, a method common to all fields of empirical knowledge, and should properly be called the empirical method ([1871] 1981: 47). The laws of theoretical economics are really never laws of nature in the true meaning of the word. On the contrary, they can be only empirical or exact laws of the ethical world ([1883] 1985: 59). Taking the methodological empiricism of, and the evolutionary element in Carl Menger’s theoretical approach as the standard, it is surely the Hayekian research program that is more entitled to claim Menger’s heritage than the Misesian praxiological paradigm. It is also the alternative Austrian economics must opt for when it wants to play a role in the discourse among the empirical social sciences.

Notes 1 Menger ([1883] 1985: 94) speaks of a “genetic” understanding of social phenomena: “Every theory … has primarily the task of teaching us to understand the concrete phenomena of the real world as exemplifications of a certain regularity in the succssion of phenomena, i.e. genetically … . This genetic element is inseparable from the idea of theoretical science.”

76

F.A. Hayek and evolutionary Austrian economics 2 Menger ([1883] 1985: 159): “The methods for the exact understanding of the origin of the ‘organically’ created social structures and those for the solution of the main problems of exact economics are by nature identical.” 3 Referring to Menger’s causal-genetic method, U. Witt and A. Chai (2019: 10, fn.) posit: “The Austrian School of economics, which Menger founded, did not adopt this method, missing the early chance to put forth a genuinely evolutionary approach.” 4 The sciences of life, mind and society, Hayek (1973, 1978: 278) comments, “are characterized by what Warren Weaver has called ‘organized complexity’ (to distinguish them from the phenomena of unorganized complexity where we can replace the information about the individual elements by statistically ascertained probabilities about the occurrence of certain elements.” Hayek refers to Weaver’s article “Science and Complexity,” American Scientist 36, 1948, 536-544. 5 Hayek (1960: 27): “(W)e command many tools – in the widest sense of that word – which the human race has evolved and which enable us to deal with our environment. … These ‘tools’ which man has evolved and which constitute such an important part of his adaption to his environment include much more than material implements. They consist in a large measure of forms of conduct which he habitually follows without knowing why; they consist of what we call ‘traditions’ and ‘institutions’.” 6 On this divide see also Vanberg 2004: 157 ff. 7 The methodological issues raised by Mises’ claims I have discussed in Vanberg 1975: 85ff. 8 For a more detailed discussion see Vanberg 2004: 183 ff.; 2017: 41 ff.

References Beck, Naomi and Ulrich Witt 2019: “Austrian Economics and the Evolutionary Paradigm,” Studies in Logic, Grammar and Rhetoric 57, 205–225. Hayek, Friedrich A. [1937] 2014a: “Economics and Knowledge,” in: The Market and Other Orders, The Collected Works of F.A. Hayek, Vol. 15, ed. by Bruce Caldwell, London and New York: Routledge, 51–77. Hayek, Friedrich A. [1967] 2014: “The Results of Human Action But Not of Human Design,” in: The Market and Other Orders. The Collected Works of F.A. Hayek, Vol. 15, ed. by Bruce Caldwell, London and New York: Routledge. 302. Hayek, Friedrich A. [1964] 2014: “The Theory of Complex Phenomena,” in: The Market and Other Orders. The Collected Works of F.A. Hayek, Vol. 15, ed. by Bruce Caldwell, London and New York: Routledge, 157–277. Hayek, Friedrich A. [1944] 2014b: “The Use of Knowledge in Society,” in: The Market and Other Orders, The Collected Works of F.A. Hayek, Vol. 15, ed. by Bruce Caldwell, London and New York: Routledge, 93–104. Hayek, Friedrich A. [1952] 2010: Studies in the Abuse and Decline of Reason, The Collected Works of F.A. Hayek, Vol. XIII, ed. by Bruce Caldwell, London and New York: Routledge. Hayek, Friedrich A. [1952] 2017: The Sensory Order and Other Writings on the Foundations of Theoretical Psychology, The Collected Works of F.A. Hayek, Vol. 14, ed. by Viktor J. Vanberg, Chicago: The University of Chicago Press. Hayek, Friedrich A. 1960: The Constitution of Liberty, Chicago: The University of Chicago Press. Hayek, Friedrich A. [1967] 1978: “Dr. Bernard Mandeville,” in: New Studies in Philosophy, Politics, Economics and the History of Ideas, Chicago: The University of Chicago Press, 249–266. Hayek, Friedrich A. [1965] 2014: “Kinds of Rationalism,” in: The Market and Other Orders. The Collected Works of F.A. Hayek, Vol. 15, ed. by Bruce Caldwell, London and New York: Routledge, 39–53. Hayek, Friedrich A. 1973: Rules and Order, Vol. 1 of Law, Legislation and Liberty, London: Routledge. Hayek, Friedrich A. [1973] 1978: “The Place of Menger’s Grundsätze in the History of Economic Theory, in: New Studies in Philosophy, Politics, Economics and the History of Ideas, Chicago: The University of Chicago Press, 270–282. Hayek, Friedrich A. 1976: The Mirage of Social Justice, Vol. 2 of Law, Legislation and Liberty, London: Routledge. Hayek, Friedrich A. 1979: The Political Order of a Free People, Vol. 3 of Law, Legislation and Liberty, London: Routledge.

77

Viktor J. Vanberg Hayek, Friedrich A. 1984: “The Origins and Effects of Our Morals: A Problem for Science,” in: Chiaki Nishiama, and Kurt R. Leube (eds.), The Essence of Hayek, Stanford: Hoover Institution Press, 319–330. Hayek, Friedrich A. 1988: The Fatal Conceit – The Errors of Socialism, The Collected Works of F.A. Hayek, Vol. 1, ed. by W.W. Bartley, III, London: Routledge. Menger, Carl [1871] 1981: Principles of Economics (Translation of Grundsätze der Volkswirtschaftslehre), New York and London: New York University Press. Menger, Carl [1883] 1985: Investigations into the Method of the Social Sciences with Special Reference to Economics (Translation of Untersuchungen über die Methode der Sozialwissenschaften und der politischen Ökonomie insbesondere), New York and London: New York University Press. Menger, Carl 1892. The Origin of Money. The Economic Journal, Vol. 2, 239–255. Mises, Ludwig von [1942] 1990: “Social Science and Natural Science,” in: Money, Method, and the Market Process – Essays by Ludwig von Mises, ed. by Richard M. Ebeling, Dordrecht: Kluwer Academic Publishers, 315. Mises, Ludwig von [1944] 1990: “The Treatment of ‘Irrationality’ in the Social Sciences,” in: Money, Method, and the Market Process – Essays by Ludwig von Mises, ed. by Richard M. Ebeling, Dordrecht: Kluwer Academic Publishers, 16–36. Mises, Ludwig von 1949: Human Action – A Treatise on Economics, New Haven: Yale University Press. Mises, Ludwig von [1962] 2006: The Ultimate Foundations of Economic Science, Indianapolis: Liberty Fund. Rothbard, Murray N. 1992: The Present State of Austrian Economics, Working Paper, Ludwig von Mises Institute ( https://cdn.mises.org/The%20Present%20State%20of%20Austrian%20Economics_2.pdf). Sanz Bas, David, Juan Morillo Benutué, and Luisa Solé Moro 2020: “Carl Menger and the birth of subjective methodology in the Economic Science,” Anuario Juridico y Económio Escurialense 53, 397–424. Vanberg, Viktor 1975: Die zwei Soziologien – Individualismus und Kollektivismus in der Sozialtheorie, Tübingen: J.C.B. Mohr, Paul Siebeck. Vanberg, Viktor 1994: “Cultural Evolution, Collective Learning and Constitutional Design,” in: David Reisman (ed.), Economic Thought and Political Theory. Boston, Dordrecht, London: Kluwer, 171–204. Vanberg, Viktor 2004: “Austrian Economics, Evolutionary Psychology, and Methodological Dualism: Subjectivism Reconsidered,” in: Evolutionary Psychology and Economic Theory, Advances in Austrian Economics 7, 155–199. Vanberg, Viktor 2017: “The ‘Knowledge Problem’ as the Integrating Theme of F.A. Hayek’s Oeuvre: An Introduction to The Sensory Order,” in: F.A. Hayek, The Sensory Order and Other Writings on the Foundations of Theoretical Psychology, The Collected Works of F.A. Hayek, Vol. 14, ed. by Viktor J. Vanberg, Chicago: The University of Chicago Press, 1–111. Witt, Ulrich and Andreas Chai 2019: “Evolutionary Economics – Taking Stock of Its Progress and Emerging Challenges,” in: idem (eds.), Understanding Economic Change – Advances in Evolutionary Economics, Cambridge: Cambridge University Press, 1–40.

78

5 KENNETH BOULDING’S CONTRIBUTION TO EVOLUTIONARY ECONOMICS Stefan Kesting

5.1

Introduction

Kenneth E. Boulding is certainly regarded as one of the fathers of evolutionary economics (Dopfer, 1994). However, did he and does he still have an influence on shaping evolutionary economics and if so in what way? This chapter will try to answer this question while also explaining what influenced the key elements of his evolutionary economics: his specific theory of economic progress, production, market exchange and government intervention?

5.2 What is Boulding’s relevance as a founding father of evolutionary economics? There certainly is no Boulding sub-school within evolutionary economics. This is probably due to his particularly holistic and idiosyncratic evolutionary theories which are not used much in the field. A full text search for “Boulding” for the years after 2000 without any further limiting selection criteria in the Journal of Economic Issues shows more than 50 articles. However, only three of these articles refer to Boulding’s work in the abstract and only one (Valentinov, 2015) is explicitly building on Boulding’s evolutionary economics. A similar search in the Journal of Evolutionary Economics brought scarcely more than twenty hits. However, these articles as such suggest an ongoing interest in at least referring to Boulding in passing by contemporary evolutionary economists. Most of these articles (nine) deal with new or alternative approaches to microeconomics in the broadest sense, ranging from game theory (Hodgson and Huang, 2012) and theories of selection (Knudsen, 2002 and Dickson, 2003) to intentionality (Muñoz and Encinar, 2014) and a psychological theory for evolutionary economics (Markey-Towler, 2018) to novelty and agents of change (Gerschlager, 2012), to the knowledge economy (Gürpınar, 2016), human capital theory (Cañibano and Potts, 2019) and the micro-meso-macro nexus (Dopfer, Foster and Potts, 2004). Another two sets of three articles focus on aspects of ecological economics such as: the circular economy (Chizaryfard, Trucco and Nuur, 2021) evolutionary thinking in environmental economics (van den Bergh, 2007) and co-evolution (van den Bergh and Stagl, 2003) and on macroeconomic (Foster, 2011), policy (Brenner and Broekel, 2019) or history

DOI: 10.4324/9780429398971-7

79

Stefan Kesting

(Leviaux and Parent, 2018). This broad range of references to Boulding’s work is due to the wide spectrum of his interests and to his general and unifying approach to the social sciences (Dolfsma and Kesting, 2013, pp. 63–109). This makes him the perfect crown witness to call upon in support of one’s own trans- and interdisciplinary research endeavours. Another important piece of evidence for Boulding’s important role as a founder is certainly Kurt Dopfer’s article “Kenneth Boulding: A Founder of Evolutionary Economics,” published in the Journal of Economic Issues in 1994. Moreover, Boulding himself defined and circumscribed Evolutionary Economics as a subject in 1991 in a contribution to the first issue of the Journal of Evolutionary Economics (Boulding, 1991). He also published two books displaying him as an evolutionary thinker who is an economist by profession (1978 and 1981). So, he definitely must be seen as a classic author in Evolutionary Economics while his founding contribution to it is not mentioned in most of the obituaries and portraits (Harcourt, 1983, Khalil, 1994, Rapport, 1996, Rapoport 1997 and Solo, 1994) apart from one (Mott, 2000). Mott also discusses evolutionary economics with Boulding at length in an interview spanning all of his contributions to economics (Mott, 1992).

5.3

What evolves in Boulding’s evolutionary economics?

While Khalil claims that “… Boulding is not an evolutionary economist, but rather an ecodynamicist” (1996, p. 86), and critiques: “The vision is not evolutionary for the simple reason that it lacks the identification of the unit of evolution” (Khalil, 1996, p. 93), Waters defends Boulding: “What evolves in Evolutionary Economics is knowledge. The increase in knowledge is developed through human processes of discovery, invention, innovation, and entrepreneurship” (2006, p. 469). Moreover, Patalano stresses the cultural acquisition of knowledge based on experience and skill development in Boulding’s conception of processes of learning: “The attention that Boulding pays to learning is symptomatic of his general interest in the process of economic change and its still unclear mechanisms” (2013, p. 393). So, though he admitted the underlying mechanism needs to be further explored, Boulding placed the learning process at the heart of human evolution: Evolution is a learning process and learning is an evolutionary process. There are processes by which more and more complex and improbable structures are created. The key concept of science is information, not matter or energy; and information is the key to evolutionary theory. Information is the only thing which is not conserved, the way matter and energy are. When a teacher teaches a class, the class knows more at the end of it and the teacher knows more too. (Boulding, 1967, p. 7) 1 Khalil’s second critical point is: “While he allows knowledge to change through learning, there is no criterion for its leaning … or for its progress” (Khalil, 1996, p. 92). However, as Robert Scott puts it: “Boulding saw himself as a modern political philosopher who was primarily concerned with the well-being of people (humanomics)” (2015, p. 185).2 Boulding labels his criterion for well-being and progress through learning: “human betterment”. Acknowledging the unavoidable vagueness of this welfare criterion, he defines it as: “By betterment I mean a process through time in which in terms of some human valuations the state of the system later in time is evaluated as superior or ‘better’ than the same system earlier in time” (Boulding, 1984, p. 1). 80

Kenneth Boulding’s contribution to evolutionary economics

This is his term for social welfare improvements and while Khalil further critiques: “It simply cannot explain the change of the fundamental technological/institutional regime of production” (Khalil, 1996, p. 97), I argued elsewhere that building on Boulding’s ideas we should use human discourse to explain the human valuation processes that lead to institutional change aiming at welfare enhancement (Kesting, 2008 and 2010). This discovery process of human betterment is based on deliberative democracy and grounded in Boulding’s concept of “the image” (1956). Warren Samuels defines it succinctly: The fundamental role of the image is to define the world. The image is the basic, final, fundamental, controlling element in all perception and thought. It largely governs our definition of reality, substantively and normatively, in part as to what is actual and what is possible. (Samuels, 1997, p. 311) The image provides an alternative theoretical foundation to explain economic decision making and behaviour and replaces the standard microeconomics of maximizing given and stable preferences within equally given constraints. Boulding describes the deliberative democratic process of attaining mutual understanding about human betterment as founded on communication: “Every decision involves the evaluation of our images of alternative futures. We also indulge in what might be called “conversational evaluation”, which does not necessarily involve decision, in which we talk either to ourselves or to others about whether “this” is better than “that”. “This” and “that” can be works of art, music, scenery, personalities, clothing, the whole state of ourselves, our family, our neighbourhood, our city, our country, the organizations we belong to, or even the whole state of the world. Evaluation is almost universally multidimensional. It involves comparisons of complex structures, rather like what an accountant calls a “position statement,” in which we put some sort of valuation on parts and then integrate these by quite complex processes into an evaluation of the whole” (Boulding, 1991a, p. 61). Boulding pointed out that these evaluations will of course differ from person to person as well as culture to culture and this can and will most likely lead to conflicts. However, he did not see conflicts as productive, but as obstacles in the evolutionary process aiming at accumulation of knowledge: “Conflict and struggle tend to be interruptions in the larger processes of maturation and learning” (1991a, p. 75). Instead he saw evolutionary progress coming “… from the rise of nonthreat organizers in society, either through exchange and the market or through integrative structures that are relaxed and tolerant, gentle and liberal” (1991a, p. 75). Reading and teaching Veblen’s work seems to have influenced Boulding’s notion of evolutionary progress and deliberative welfare economics significantly as noted by Babe (1995, pp. 198, 199); Scott (2015, p. 135)3 and Waller (2013, p. 267). Robert Heilbroner raised two critical points early on while reviewing Boulding’s work up to 1975: firstly, he found: “Boulding treats power as an abstract quality of social systems rather than as the specific capability of persons and institutions to impose their wills in historically defined, and therefore changeable, circumstances” (1975, p. 78). Secondly, Heilbroner observes that Boulding’s theoretical generalizations are too lofty, not grounded enough in reality: “… political economy in Boulding’s hands becomes an a-historical – even an antihistorical – instrument” (1975, p. 79). Heilbroner’s critical observations still strike me as valid even though Boulding used and promoted the historical method repeatedly in his work (Boulding, 1981, pp. 123–146, 1988, see the charts in the appendix, 1992, 1993) and 81

Stefan Kesting

seems to have tried to address these two critical points by writing a whole book on power (1989) and another monograph on the economic history of the USA (1993). There is definitely an idealistic tone and utopian streak in his concepts of evolutionary progress and a holistic bird’s eye perspective in Boulding’s interpretation of history.4

5.4

The key influences on and elements of Boulding’s evolutionary economics

His welfare theory aside, let’s try to tackle the question of what the key elements of Boulding’s evolutionary economic theories are. Apart from Veblen and Darwin, I see three major theoretical influences on his evolutionary method of inquiry: his interest in biology, as well as the influence of Schumpeter and Keynes. Heilbroner observed “What Boulding seeks throughout his work is a higher level of systemic analysis: As with Alfred Marshall,5 his Mecca lies in ’economic biology’” (Heilbroner, 1975, p. 73). Boulding’s interest in the biological implications of the population concept and genetics led him to view social evolution in analogy to biological evolution6 and to focus on the Darwinian concept of the niche. Waters7 summarises this approach very concisely: Boulding describes an ecosystem as composed of innumerable niches for different kinds of creatures and behaviors. A niche is the potential equilibrium population of a species in the ecosystem (Boulding, 1981, p. 31). The driving force of Boulding’s systems perspective is that evolution is a process of change in genetic structure, which he identifies as mutation. He believes the basic evolutionary process is the accumulation of knowledge, i.e. changes in genetic structure of species. By genetic structure he meant any egg, design, or plan that contains the instruction for producing a phenotype such as a chicken, university or building. … As he indicated above, biological and societal evolution consists mainly in filling of empty niches in the course of mutation and selection. Societal, hence economic, mutation results from invention, discovery, and other creative activities of humans. Whereas biological changes in DNA are largely random, often stimulated by environmental stress, societal mutations are often intentional efforts of people to create something new. The motive force may be societal or environmental stress, self-interest, curiosity, or accident. In biological selection, the members of the species that best fit the niche in terms of survival tend to fill the niche, given the environmental conditions. Societal selection is quite similar. (Waters, 2013, p.Waters Dolfsma 2013 289) I agree with Valentinov (2015, p. 85) observing that this biology inspired Darwinian ecological approach probably hampered Boulding’s influence on evolutionary economics.

5.5

Theory of production

As Boulding states in an interview with Tracy Mott: Production is how you get from the genotype to the phenotype, whether this is the automobile or the horse. It is very similar. Except that the genotype of the automobile isn’t in the automobile. It is in all sorts of people’s minds, human brains, plans, computers, and what not. But you start off with a genotype which is in a sense a plan. (Mott, 1992, p. 350) 82

Kenneth Boulding’s contribution to evolutionary economics

So, Boulding’s theory of production is a translation or application of biological theories of genetics and evolution to the socio-economic sphere while combining it with his image based theory of evolutionary progress in knowledge. However, his model of production founded on this two-fold conceptual foundation is set in the context of power in chapter 11, entitled Power in Society of his book Ecodynamics: “Economic power, that is, command over economic goods, comes from a number of different sources. It comes mainly from control over the factors of production8 that enter into the process of production of economic goods. As we have seen earlier, these consist essentially of knowledge or know-how, energy, and materials. The greatest of these is knowledge, although there are occasions in which the power to produce economic goods is limited by lack of access to energy or to materials” (1978, p. 243). His production model presents a rare occasion where Boulding proposes an equation to measure the economic power of a population as output of economic goods per capita real income. A group of people L does either produce for domestic use (Ld), for exchange (Le) or does not produce while enjoying leisure, sleeping or because of being unemployed (Lu). So, L = Ld +Le + Lu. Based on this, Boulding continues: If pd is the productivity of domestic activity, pe the productivity of production for export, T the terms of trade (that is, how many imports are obtained per unit of exports), and G the total of grants, gifts, or transfers (which may of course be positive or negative),9 the total real income Y is: Y = Ldpd + LepeT + G. Then, we have the per capita real income or economic power: y=

Ld pd + Le pe T + G Y = L Ld + Le + L u

This is an instructive equation, showing the sources of an increase in economic power. A shift out of Lu, that is, unemployed potential, into either Ld or Le will increase economic power: y. Shifts into production for export out of domestic production will increase only if pe is greater than pd. An increase in positive grants, of course, will increase economic power and so will an improvement in the terms of trade, T, … . The major long-run source of increase in economic power, however, is increase in productivity, that is, in pd or pe. This is the only source that can be sustained for very long. (1978, p. 244) Note that what anthropologists or sociologists would call “gifts” (i.e. voluntary unpaid work, financial contribution to charity etc.) is not seen as an unproductive form of redistribution of income by Boulding, but as a productive contributing factor to overall welfare, to enhance income. Moreover, Boulding’s theory of production is a clear expression of his conviction that knowledge accumulation drives progress in economic evolution: “Increase in productivity often comes from education, training, or an increase in skill” (1978, p. 244). I find myself wondering how a production function based on Boulding’s theory would look like, but came to no satisfactory conclusion. However, drawing an illustrative graph Waters provides an excellent overview of Boulding’s approach to production as the key driver of the process of societal evolution (2013, p. 295).

83

Stefan Kesting

5.6 Theory of the market Boulding writes with great appreciation of exchange and the market as a selection mechanism: “The great virtue of the market is that it economizes agreement” (1981, p. 178). However, he is rejecting the standard model of the market which he sees as too simplistic to have any explanatory value: “The theory of maximizing behaviour and the marginal analysis is essentially “Newtonian”. It is a mechanical, not an evolutionary model” (1981, p. 99) and states: “Equilibrium in such a system is a useful intellectual construct, even though it is never found in the real world” (1981, p. 87). Inspired by Schumpeter (Boulding, 1981, p. 85) Boulding’s understanding of the market is a sophisticated Austrian10 one: “A market-type economy by contrast is much more like an ecosystem, in which all the various commodities and social products of all kinds interact with each other ecologically and survive if they can find a niche in the total system” (Boulding, 1981, p. 41). Moreover, he asks: When everybody is supposed to maximize profits, how is it that so many organizations suffer losses or even become bankrupt? The answer is, of course, that … their image of themselves, the environment and the world around them, and the future – may mutate in ways that do not lead to survival. This happens all the time in biological evolution: if a biogenetic mutation leads to survival it is simply good luck; most such mutations are adverse. (1981, p. 103) Survival is a matter of finding an empty niche: “The corner grocery still survives in the cracks of the supermarket, the hippies in old mining towns, pacifists in churches, radicals in universities, families in leisure time, and so on” (1981, p. 74). Moreover, “… an evolutionary perspective also recognizes that there are advantages to preserving some redundancy to enable sufficient adaptability to changes in external conditions” (Mott, 2000, p. F439). In Boulding’s view, market exchange is only one among a number of alternative selection mechanisms. As Valentinov observes: However, exchange relationships, while accommodating substantial complexity, fail to “create community, identity, and commitment … This, indeed, is one of the great weaknesses of capitalism, which is organized principally through exchange. It may not be able to attract through its institutions that minimum of loyalty, devotion, and affection necessary to maintain them. Joseph A. Schumpeter was perhaps the first economist to point this out” (Boulding, 1981a, p. 33). The role of the integrative system is in delivering precisely this type of legitimacy through “community, identity, and commitment” that was of concern to Schumpeter (1993) (Valentinov, 2015, p. 74).

5.7

Government intervention

As Scott highlights: “Boulding was for ever Keynesian” (2015, p. 33). Not only is the second edition of Boulding’s textbook Economic Analysis (1948) “entirely Keynesian – adopting concepts from The General Theory” (Scott, 2015, p. 46), but there are also whole sections in chapter 3: Evolutionary and “Mainline” Economics (pp. 112–121) in Evolutionary Economics

84

Kenneth Boulding’s contribution to evolutionary economics

(1981) discussing Keynesian ideas favourably. Moreover, Boulding stressed the positive role of government in enhancing general welfare and correcting market failure (1981, p. 171). Scott summarises Boulding’s views: Keynes’s economics, however, results in many benefits for society. Keynes’s policies after the Great Depression in the United States ushered in a time of economic strength with a good balance between government and market. Keynes created macroeconomics, which gave rise to economic accounting that made it possible to understand business cycles, widespread unemployment, aggregate demand shortages, and money. (2015, p. 146). In his combination of a Schumpeterian cum Keynesian perspective Boulding concludes: The sociological dilemma of capitalism lies in the fact that it destroys, and must destroy, the community of ascription, and it may find the community of achievement difficult to establish. The danger, therefore, is that capitalism may destroy the community altogether, and leave us with a world of isolated, communityless, and thereby almost of necessity detached, dangerous and neurotic individuals. (Boulding, 1971–1973, vol. II, p. 79) So, Heilbroner is certainly right when he states: “I think Boulding would not object to being classified as a libertarian socialist” (1975, p. 76). For most economists to hold such politically contradictory positions at the same time seems probably impossible, but not for Boulding who possessed a mind of enormous intellectual breadth and tolerance.

Notes 1 I found this quote in Robert Scott’s wonderful book ( 2015, p. 126) which reviews Boulding’s contribution to Quaker events and publications – an otherwise mostly neglected aspect of Boulding’s work. 2 Deirdre McCloskey also sees Boulding as contributing to what she calls “humanomics”: “We think humanomics is the future of a truly scientific economics. Pounding away endlessly on the Samuelsonian formula of Max U s.t. C reduces everything, simply everything, to the virtue of Prudence and its corresponding vice Greed. It is time to get serious about human motivation, … . Boulding was in fact a pioneer of humanomics” ( 2013, pp. 580, 581). 3 “… the fact that much of Boulding’s writing during this time (1970s) and beyond drew heavily from Veblen’s writing on social structures. Boulding’s evolutionary thinking begins from a proposition that people construct institutions that in turn affect society. It is decision making within institutions that enacts change, which is precisely in-line with institutional economics” ( Scott, 2015, p. 135). 4 Heilbroner’s methodological critique: “It is a continuing effort to formulate a theory of social continuity, an interacting ensemble of bonds and constraints capable of explaining or elucidating the cohesive properties of social systems taken in the large” ( 1975, pp. 75, 76). Moreover: “… his enlarged vision of a social system may not yield “researchable hypotheses.” “The diagrams of populations in competition or complementarity give us vivid gestalts, but do not seem to advance the practical state of the art: We do not know what to do with the insights” ( 1975, p. 77). However, these charts where also an important instrument of Boulding’s pedagogy ( Fontaine, 2021). 5 Boulding refers to Marshall as a fellow evolutionary economist ( 1981, p. 84). 6 He discusses seven variants of ecological interaction: mutual cooperation, parasitism, predation, mutual competition, dominant-cooperative, dominant-competitive and mutual independence (isolated habitats) in chapter 4 on ecological dynamics in his Ecodynamics ( 1978, pp. 78–82).

85

Stefan Kesting 7 I quote Waters (2013) here at length because he provides the best short introduction to Boulding’s evolutionary economics I am aware of. 8 This “control over the factors of production” may sound Marxist. However, interviewed by Tracy Mott, Boulding states clearly: “My attitude about Marxism, as you know, has been rather unfavourable, …” ( Mott, 1992, p. 358). Moreover, his theory of production is explicitly formulated in opposition to any theories of surplus value of labour: “It is not “labor” that produces a commodity or product, as Marx and indeed Adam Smith and Ricardo thought, but human knowledge and know-how, operating through institutions which enable this know-how to capture energy and rearrange materials. Labor, in the sense of what is bought with wages, is a highly variable mixture of the three real factors of production” ( Boulding, 1981, p. 186). 9 Boulding’s Grants Economics ( 1981a) is very much inspired by economic anthropology and sociology. However, he adds his very own and very original twist to his economics of gift-giving and tributes. For more on his theory and an application of it to the Marshall Plan, see my book chapter ( Kesting, 2021). 10 Marmefelt (2009) shows parallels between Boulding and Hayek in the way they approach social evolution in their theories. Which is not surprising given the central role knowledge and innovation plays in the evolving progress in the economy for both of these economists.

References Babe, R. E. 1995. Commentary article: The communication theory of Kenneth E. Boulding. In: Dolfsma, W. and Kesting, S. Eds. Interdisciplinary Economics – Kenneth E. Boulding’s engagement in the sciences. London and New York: Routledge, pp. 184–202. Boulding, K. E. 1948. Economic Analysis. New York: Harper & Brothers. Boulding, K. E. 1956. The Image. Ann Arbor: The University of Michigan Press. Boulding, K. E. 1967. Mayer/Boulding Dialogue on Peace Research. Pendle Hill Pamphlet No. 153. Wallingford, PA: Pendle Hill Publications. Boulding, K. E. 1971–1973. Collected Papers. Three volumes. Boulder: Colorado Associated University Press. Boulding, K. E. 1978. Ecodynamics – A New Theory of Societal Evolution. Beverly Hills and London: Sage Publications. Boulding, K. E. 1981. Evolutionary Economics. Beverly Hills and London: Sage Publications. Boulding, K. E. 1981a. A Preface to Grants Economics: The Economy of Love and Fear. New York: Praeger. Boulding, K. E. 1984. How do things go from bad to better? The contribution of economics. In: Boulding, K. E. Ed. The Economics of Human Betterment. Beverly Hills: Sage or New York: Albany State University of New York Press, pp. 1–14. Boulding, K. E. 1988. What Do We Want in an Economics Textbook? The Journal of Economic Education. 19(2), pp. 113–132. Boulding, K. E. 1989. The Three Faces of Power. Newbury Park, CA: Sage Publications. Boulding, K. E. 1991. What is evolutionary economics? Journal of Evolutionary Economics. 1, pp. 9–17. Boulding, K. E. 1991a. Power and Betterment in the Economy. In: Boulding, E. and Boulding, K. E. Eds. 1995. The Future – Images and Processes. Thousand Oaks, London and New Delhi: Sage Publications, pp. 57–75. Boulding, K. E. 1992. Appropriate methodologies for the study of the economy. In: Boulding, K. E. Ed. Towards a New Economics. Aldershot: Edward Elgar, pp. 98–111. Boulding, K. E. 1993. The Structure of a Modern Economy: The United States, 1929–1989. New York: New York University Press. Brenner, T. and Broekel, T. 2019. Evolutionary economics and policy: Introduction to the special issue. Journal of Evolutionary Economics. 29, pp. 1373–1378. 10.1007/s00191-019-00646-7 Cañibano, C. and Potts, J. 2019. Toward an evolutionary theory of human capital. Journal of Evolutionary Economics. 29, pp. 1017–1035. 10.1007/s00191-018-0588-y Chizaryfard, A., Trucco, P. and Nuur, C. 2021. The transformation to a circular economy: Framing an evolutionary view. Journal of Evolutionary Economics. 31, pp.: 475–504. 10.1007/s00191-020-00709-0

86

Kenneth Boulding’s contribution to evolutionary economics Dickson, P. 2003. The pigeon breeders’ cup a selection theory of economic evolution. Journal of Evolutionary Economics. 13, pp. 259–280. 10.1007/s00191-003-0151-2 Dolfsma, W. and Kesting, S. 2013. Interdisciplinary Economics – Kenneth E. Boulding’s engagement in the sciences. London and New York: Routledge. Dopfer, K. 1994. Kenneth Boulding: A founder of Evolutionary Economics. Journal of Economic Issues. 20(4), pp. 1201–1204. Dopfer, K., Foster, J. and Potts, J. 2004. Micro–meso–macro. Journal of Evolutionary Economics. 14, pp. 263–279. 10.1007/s00191-004-0193-0 Fontaine, P. 2021. Kenneth Boulding’s visual imagination: From textbooks to scholarship, 1941–62. History of Political Economy. 53(2), pp. 213–241. 10.1215/00182702-8905991 Foster, J. 2011. Evolutionary macroeconomics: A research agenda. Journal of Evolutionary Economics. 21, pp. 5–28. 10.1007/s00191-010-0187-z Gerschlager, C. 2012. Agents of Change. Journal of Evolutionary Economics. 22, pp. 413–441. 10.1007/ s00191-011-0262-0 Gürpınar, E. 2016. Organizational forms in the knowledge economy: A comparative institutional analysis. Journal of Evolutionary Economics. 26, pp. 501–518. DOI 10.1007/s00191-016-0452-x Harcourt, G. 1983. A man of all systems: talking with Kenneth Boulding. Journal of Post Keynesian Economics. 6(1), pp. 143–154. Heilbroner, R. 1975. Kenneth Boulding, “Collected Papers”: A review article. Journal of Economic Issues. 9(1), pp. 73–79. Hodgson, G. M., and Huang, K. 2012. Evolutionary game theory and evolutionary economics are they different species? Journal of Evolutionary Economics. 22, pp. 345–366. 10.1007/s00191-010-0203-3 Kesting, S. 2008. Communication in the economy: The example of innovation. In: Davis, J. B. and Dolfsma, W. Eds. Elgar Companion to Social Economics, Cheltenham UK and Northampton US: Edward Elgar, pp. 406–426. Kesting, S. 2010. Boulding’s welfare approach of communicative deliberation. Ecological Economics. 69, pp. 973–977. Kesting, S. 2021. The fluid nature of gifts and grants – An institutional application to the Marshall Plan. Chapter 7. In: Kesting, S., Negru, I. and Silvestri, P. Eds. The Gift in the Economy and Society: Perspectives from Institutional Economics and Other Social Sciences. London and New York: Routledge, pp. 121–140. Khalil, E. 1994. Kenneth Boulding, 1910–1993. Journal of Economic Methodology. 1(1), pp. 161–166. Khalil, E. 1996. Kenneth Boulding: Ecodynamicist or evolutionary economist? Journal of Post Keynesian Economics. 19(1), pp. 83–100. Knudsen, T. 2002. Economic Selection Theory. Journal of Evolutionary Economics. 12, pp. 443–470. Leviaux, P. and Parent, A. 2018. The biological hypothesis in cliometrics of growth: A methodological critique of Fogel (post 1982) and Ashraf & Galor (2013). Journal of Evolutionary Economics. 28, pp. 929–950. 10.1007/s00191-018-0560-x Markey-Towler, B. 2018. A formal psychological theory for evolutionary economics. Journal of Evolutionary Economics. 28, pp. 691–725. 10.1007/s00191-018-0566-4 Marmefelt, T. 2009. Human knowledge, rules, and the spontaneous evolution of society in the social thought of Darwin, Hayek, and Boulding. Journal of Economic Behavior & Organization. 71, pp. 62–74. 10.1016/j.jebo.2009.02.013 McCloskey, D. 2013. Comment: What went wrong with economics – A quarter century on. In: Dolfsma, W. and Kesting, S. Eds. Interdisciplinary Economics – Kenneth E. Boulding’s engagement in the sciences. London and New York: Routledge, pp. 574–586. Mott, T. 1992. Kenneth Boulding interviewed by Tracy Mott 28–29 March 1991. Review of Political Economy. 4(3), pp. 341–374. Mott, T. 2000. Kenneth Boulding, 1910–1993. The Economic Journal. 110(June), pp. F430–F444. Muñoz, F.-F. and Encinar, M.-I. 2014. Intentionality and the emergence of complexity: An analytical approach. Journal of Evolutionary Economics. 24, pp. 317–334. 10.1007/s00191-014-0342-z Patalano, R. 2013. Comment: Culture, mind and context – The revolutionary road of Kenneth E. Boulding. In: Dolfsma, W. and Kesting, S. Eds. Interdisciplinary Economics – Kenneth E. Boulding’s engagement in the sciences. London and New York: Routledge, pp. 384–401. Rapoport, A. 1997. Memories of Kenneth E. Boulding. Review of Social Economics. 55(4), pp. 416–431.

87

Stefan Kesting Rapport, D. 1996. In memory of Kenneth E. Boulding. Ecological Economics. 17, pp. 67–71. Samuels, W. J. 1997. Kenneth Boulding’s the image and contemporary discourse analysis. In: Samuels, W. J., Medema, S. G. and Schmid, A. A. Eds. The Economy as a Process of Valuation. Cheltenham: Edward Elgar, pp. 299–327. Scott, R. 2015. Kenneth Boulding: A Voice Crying in the Wilderness. Houndmills, Basingstoke: Palgrave Macmillan. Schumpeter, J. A. 1993. Kapitalismus, Sozialismus und Demokratie. Tübingen: Francke Verlag. Solo, R. 1994. Kenneth Ewart Boulding: 1910–1993. An appreciation. Journal of Economic Issues. 28(4), pp. 1187–1200. Valentinov, V. 2015. Kenneth Boulding’s Theories of Evolutionary Economics and Organizational Change: A reconstruction. Journal of Economic Issues. 49(1), pp. 71–88. 10.1080/00213624.2015. 1013880 van den Bergh, J. C. J. M. 2007. Evolutionary thinking in environmental economics. Journal of Evolutionary Economics. 17, pp. 521–549. 10.1007/s00191-006-0054-0 van den Bergh, J. C. J. M. and Stagl, S. 2003. Coevolution of economic behaviour and institutions: Towards a theory of institutional change. Journal of Evolutionary Economics. 13, pp. 289–317. 10.1007/ s00191-003-0158-8 Waller, W. 2013. Comment: Kenneth Boulding on power. In: Dolfsma, W. and Kesting, S. Eds. Interdisciplinary Economics – Kenneth E. Boulding’s engagement in the sciences. London and New York:Routledge, pp. 258–268. Waters, R. 2006. What Happened to Boulding’s Evolutionary Economics? Journal of Economic Issues. 40(2), pp. 465–471. Waters, R. 2013. Comment: Evolutionary Economics – A framework for organizational decisionmaking. In: Dolfsma, W. and Kesting, S. Eds. Interdisciplinary Economics – Kenneth E. Boulding’s engagement in the sciences. London and New York: Routledge, pp. 287–301.

88

6 EVOLUTIONARY ECONOMICS AND PSYCHOLOGY Where we are, where we could go Brendan Markey-Towler

6.1

Introduction: A good foundation for further building

Among the schools of economic thought, even, to some extent, behavioural economics, evolutionary economics is arguably the most consistent with psychology, and the one that most integrates psychology into its analytical foundations. The reason for this is at least partly historical. Evolutionary economics co-developed with various movements in psychology and incorporated much of those into itself. While this means evolutionary economics has the advantage of strong consistency with the science of human behaviour, there is more that can be done to expand the present psychological foundations for evolutionary economics. In this chapter, we discuss psychological foundations of the broad sweep of traditions in evolutionary economics. We discuss how the neo-Schumpeterian, Veblenian and “Naturalistic”, or “bioeconomic” perspectives are generally consistent with and have a foundation in the cognitive tradition within psychology, as well as the evolutionary tradition broadly speaking. We then outline opportunities for growth that could be realised by integrating a selection of movements within psychology into the analytical base of evolutionary economics. We discuss four subfields evolutionary economics could benefit from: personality, social, affective and persuasion psychology. We conclude by surveying the current psychological foundations of evolutionary economics, and by a call to further integration with psychology proper that expands the analytical foundations of evolutionary economics. We do not pretend to embark on a thorough literature review, preferring instead to draw together high-level observations of the evolutionary economics literature. We also, perhaps surprisingly, do not discuss the potential for behavioural economics to integrate within evolutionary economics. This field, as can be roughly represented by the works of Daniel Kahneman and Amos Tversky as well as Richard Thaler (Kahneman, 2003, 2011; Thaler, 2015) tends to draw on empirical findings especially from cognitive and social psychology to critique neoclassical models. Our preference is instead to indicate paths for evolutionary economics to move to a deeper integration with psychology proper.

DOI: 10.4324/9780429398971-8

89

Brendan Markey-Towler

6.2

Existing foundations: Neo-Schumpeterian, Veblenian and naturalistic perspectives

Within evolutionary economics we can identity two strong analytical traditions, and an emergent tradition that provide the foundation most research in the field. The first two fields are the neo-Schumpeterian tradition descending from the work of Richard Nelson and Sidney Winter, and the Veblenian perspective descending from Thorstein Veblen, curated by Geoffrey Hodgson. The third tradition, which we might call the “Naturalistic” or “Bioeconomic” perspective, is emergent from the works of Ulrich Witt and Hermann Carsten-Pillath seeking to integrate natural science foundations into evolutionary economics. Each of these traditions substantially informs the “micro-meso-macro” synthesis developed by Kurt Dopfer, John Foster and Jason Potts (Dopfer, Foster and Potts, 2004; Dopfer and Potts, 2007), and certainly its psychological foundation (Dopfer, 2004). Revisiting the analytical framework of these traditions as outlined in their core texts, we may see that evolutionary economics has a strong foundation in particularly the cognitive and evolutionary traditions of psychology, and thus has a strong consistency with the science of human behaviour. This is partly a result of history: these traditions co-developed with major movements in cognitive and evolutionary psychology. We can then see how this solid grounding in psychology provides us with a foundation to realise opportunities for growth by integrating further psychological movements. As the major, summative text of the cognitive and evolutionary psychology movement, we take the Steven Pinker’s How the Mind Works (1997). This text draws together the cognitive tradition descended particularly from the works of Alan Newell (see Newell, 1990) as well as the evolutionary tradition systematised particularly by Leda Cosmides and John Tooby (see Cosmides and Tooby, 2013). The cognitive movement views the mind as akin to, or even as a biological form of, a computer. The mind is understood as something more than analogous to an operating system containing a variety of algorithms using simple logical operators (AND/OR/IF-THEN) to encode information processing capabilities. These simple operators chain together into an immensely complex system that achieves the feat of transforming base sensory information into higher-order abstractions that can form the basis for sophisticated models that allow us to live and work effectively in the world. The evolutionary psychology movement can be thought to place this (as well as other traditions) within an evolutionary perspective whereby psychological traits interact within selection pressures exerted by differential survival and reproduction. Over time, psychological traits can become adapted to more effective dispatch of tasks that improve survival and reproduction capability by the differential selection of their carriers.

6.3 Neo-Schumpeterian perspectives: Cognitive psychology The neo-Schumpeterian tradition begins with the publication in 1982 of Richard Nelson and Sidney Winter’s monumental text An Evolutionary Theory of Economic Change (Nelson and Winter, 1982). In this text, Nelson and Winter expanded upon the works of Joseph Schumpeter, who introduced the argument that the economy was consistently thrown out of equilibrium into a process he called “development” by the activity of entrepreneurs and innovators introducing new “combinations” of resources (Schumpeter, 1911, 1942). Nelson and Winter took this perspective and transformed it into an evolutionary model of economic change through technological advance and adoption.

90

Evolutionary economics and psychology

In Nelson and Winter’s model, evolutionary pressures are exerted on firms by their differential survival in the marketplace. Firms that are selected and retained are those which successfully accrue customers and profits. Their capability for selection and retention emerges from the interaction between their internal production routines and demand in the market for the outputs produced by those routines. Technological advance manifests in new routines that introduce variations to these routines and improve the capability for selection and retention by those firms which drove such advance through innovation and entrepreneurship. Over time, market demand can be expected to converge on only those firms whose routines are best adapted to the market’s conditions. While developing their model, Nelson and Winter drew explicitly on the emerging field of organisational psychology. In particular, they drew on the behavioural theory of the firm developed by the followers of Herbert Simon in the wake of his 1947 book Administrative Behavior (Simon, 1947). Simon himself in partnership with James March, and then later March and Richard Cyert, developed a perspective on firms as sets of routines for transforming information and resources into outputs and prices (March and Simon, 1958; Cyert and March, 1963). These routines had their grounding and hold over employee behaviour in the rules, or algorithms, that governed the human mind and behaviour in that context. In other words, firms were emergent cognitive operating systems for transforming information and resources into outputs. What Nelson and Winter had done therefore, was to place this cognitive perspective on economic behaviour within an evolutionary economic context, where the cognitive traits of firms were subject to evolutionary pressures exerted by the market. The field that they were drawing on grew and evolved into the cognitive psychology tradition particularly through the work of Herbert Simon’s student Alan Newell, who together with Simon would generalise the algorithmic view of the mind beyond the firm to generalised problem solving and behaviour (Newell and Simon, 1972; Newell, 1990). Nelson and Winter’s model pre-empted later developments in evolutionary psychology insofar as they placed the cognitive traits of firms within a broader evolutionary context where those traits were subject to evolutionary pressure by the differential survival of their carriers. Nelson and Winter had thus set the analytical foundations of neo-Schumpeterian evolutionary economics firmly in cognitive psychology and integrated that within a perspective relatively isomorphic with evolutionary psychology.

6.4

Veblenian perspectives: Habit psychology

The Veblenian tradition in evolutionary economics originates in the works of Thorstein Veblen around the turn of the 20th century but went into an extended hibernation during the ascendency of the neoclassical school (Hodgson, 2004). Veblen introduced a sociological perspective on the economy as a system embedded within a broader social structure governed by social institutions – rules broadly held to across society dictating the appropriate behaviour of individuals in given social settings (Veblen, 1898, 1899, 1914). This perspective re-emerged toward the 20th century within the heterodox branch of a reinvigorated institutional economics (Hodgson, 1998, 2004), but it also substantially influenced evolutionary economics. In the Veblenian model as developed particularly by Geoffrey Hodgson, the economy emerges from the behaviour of individuals who are subject to the pressure of social institutions, which take the form of rules delineating appropriate and inappropriate behaviour 91

Brendan Markey-Towler

in given social environments. These institutions emerge from the gradual habituation of certain behavioural patterns by individuals as they go about their lives receiving feedback in their social interactions about what is and is not appropriate behaviour for them to adopt (Hodgson, 2010; Hodgson and Knudsen, 2010). Those habits may evolve as variations are introduced by individuals habituating behavioural patterns “with a twist”, and as these variations enter the population, they are subject to evolutionary pressures. If they are differentially spread and habituated across the population, they become “new” social institutions, causing the structure of society and economy to evolve with them. Hodgson draws attention to how Veblen’s perspective on the habituation of social institutions into behavioural patterns was influenced heavily by the contemporaneous psychology of William James. James, one of the fathers of modern scientific psychology, introduced habit psychology in his famous two-volume work The Principles of Psychology, where he examined how the progressive growth and strengthening of neural circuitry could be thought to underlie habit formation (James, 1890). Habits, in Jamesian psychology, are routines, or rules, of thought that begin in conscious effort to identify relations between objects and events in the environment and appropriate behavioural responses to them, and through repeated use become increasingly automated and subconscious. Veblen therefore began, and Hodgson later substantially advanced the process of placing a habit psychology perspective at the analytical core of evolutionary economics. While Jamesian psychology, strictly speaking, was somewhat eclipsed by the emergence of behaviourism, it continued to be a powerful tradition. It would substantially inform the development of cognitive and evolutionary psychology insofar as habituation across the population could be reconceptualised slightly as the gradual adaptation of cognitive routines to the environment in response to selection pressures exerted by social interactions. Thus, Veblenian evolutionary economics, in effect, took the root of cognitive and evolutionary psychology and placed it at the heart of the analytical framework of evolutionary economics.

6.5

Naturalistic and bioeconomic perspectives: Evolutionary psychology

The works of Ulrich Witt and Carsten Hermann-Pillath offer the emerging core of what might be called a “naturalistic” or “bioeconomic” perspective on evolutionary economics. These thinkers are somewhat distinguished from other traditions by their seeking not to develop evolutionary models to economic systems per se, but rather their attempt to build from natural science foundations “up” to models of economic systems. Hence, we could say this tradition is constituted by an exercise in grounding evolutionary economics in the natural sciences, particularly biology. In the naturalistic perspective, the economy is a system that emerges from the biophysical world in which human beings are subject to the pressures of natural and sexual selection to survive and reproduce (Hermann-Pillath, 2013; Witt, 1999, 2001). The economy is one system that emerges from interaction between human beings to further their objectives in survival and reproduction, and so their behavioural patterns can be understood as the result of these pressures. The imperative to which human behaviour is subject, and to which its patterns will gradually adapt, is to survive and reproduce, those traits which enhance that capability will be differentially selected and retained, and those which don’t will be discarded. Thusly we can understand the demand for and emergence of new technologies that cause the economy to evolve as an emergent function of deep evolutionary processes within 92

Evolutionary economics and psychology

the biophysical world. Demand emerges from needs that are imposed by the imperative of survival and reproduction and serves to arbitrate between alternative competition technologies which best fits perceptions of what will enhance human capability to that end. Evolutionary psychology can be thought of roughly as the study of behavioural phenotypes (genotypes being more the focus of evolutionary biology) that have emerged in response to the evolutionary pressures of selection through differential survival and reproduction. It is a subject grounded in a perspective on human behaviour as deeply embedded within the natural biophysical world, hence the title of E.O. Wilson’s seminal contribution to the field, Sociobiology (Wilson 1975), indicating the importance of understanding social interactions as an emergent biological phenomenon. In setting the economy within a naturalistic biophysical context, where human social interaction is emergent from behavioural patterns subject to evolutionary pressures, the naturalistic or bioeconomic perspective draws explicitly on this tradition within psychology. Thus, what naturalistic or bioeconomic perspectives achieve is, effectively, to place evolutionary psychology at the core of the analytical foundation of evolutionary economics.

6.6

Opportunities for growth: Personality, social, affective, and persuasion psychology

We have seen that evolutionary economics has a strong grounding in and consistency with cognitive and evolutionary psychology. Cognitive and evolutionary psychology are two major movements within psychology, and even perhaps the major integrative movements. But they are not exhaustive, and a range of other perspectives present opportunities for evolutionary economics. We will investigate the opportunities presented by four fields. Personality psychology, social psychology, affective psychology, and persuasion psychology each offer insights that may allow us to further enrich the analytical foundations of evolutionary economics and offer new insight. Each offers us a perspective on some of the factors influencing the tendency for individuals in society to adopt or resist new technologies either in their employment, or in their broader lifestyle. To obtain an initial vantage point, we will adopt the terms of Lewin’s “force field analysis”, which views any human behaviour as the outcome of “driving” and “restraining” forces (Lewin, 1938). In the context of evolutionary economics, driving forces motivate the individual toward behaviour such as the adoption of a new technology, and restraining forces create resistance on the part of the individual to adopting a new technology.

6.7

Personality psychology

The formal conceptual foundations of personality psychology were provided by George Kelly in a famous 1963 work, which introduced the idea of personal constructions of reality that construe relationships between objects, events and appropriate behaviours and thus “channelise” thought (Kelly, 1963). People tend, in Kelly’s formulation, to seek to behave in a way that is consistent with these personal constructs and suffer greatly when contradictions to them arise which cannot be incorporated into them. They constitute the expression of someone’s personality, their image of themselves and their relationship to the world. Personality psychology can be thought to comprise a study of the general tendencies within these personal constructions of reality and thus how they shape psychological reactions to an individual’s environment. The present “gold standard” is the Big 5 typology 93

Brendan Markey-Towler

(Digman, 1990; Goldberg, 1993), sometimes summarised by the acronym OCEAN. It locates personalities relative to axes indicating the strength of personality traits along five dimensions: Openness to experience, Conscientiousness in life and work activities, Extraversion in the direction of consciousness “outward” from the psyche, Agreeableness in life and work settings, and Neuroticism in susceptibility to negative emotion. Personality psychology, if integrated into evolutionary economics, could allow us greater insight into the psychological traits of innovators and entrepreneurs, and the segmentation of the population by their willingness to adopt new technologies. We can imagine that Openness, Conscientiousness, and Extraversion would tend to provide driving forces toward the adoption of new technologies, while Agreeableness and Neuroticism would generate restraining forces. This may allow us to understand better the personality traits that underlie and generate resistance to economic evolution and technological change and advance our understand of both within our analytical frameworks.

6.8

Social psychology

Social psychology has its origins in the desire to understand the extent to which individuals could resist the pressure to conform to the beliefs and behaviour of the groups that they find themselves in. Solomon Asch showed that the pressure to conform could be surprisingly strong, even to the extent of misreporting the true length of a line in a visual perception test to “fit in”, and that relatively few persons can resist pressure to conform (Asch, 1956). Much of subsequent social psychology sought to expand on this demonstration of social conformity pressure to discover its nuances and dynamics. We know, for instance, behavioural patterns can spread from person to person simply by observation and internalisation (Bandura, 1977). By observing the behaviour of another, we tend to “model” that behaviour and internalise the requisite mental states that would give rise to it, and thus we can observe behavioural patterns spread by a process of “social learning”. Alternatively, if we identify our sense of self, our emotions and our values, with our inclusion in a group, we can observe the incorporation of a “social identity” into an individual’s mind whereby the perceptions, values, and emotions of the group can influence or even determine those of the individual. Social identity theory, in other words, describes how our behavioural patterns may be more determined by the groups of which we are a part rather than our own personal characteristics (Tajfel, 1974). Integrating more social psychology into the analytical foundations of evolutionary economics may help us to understand the forces exerted by groups that effect the adoption of new technologies. In the early stages of technological adoption, we may expect many groups to create significant restraining forces inhibiting innovation, entrepreneurship, and adoption simply because they have not yet adopted it. But for early adopters, and subsequently as critical mass is gained, groups will increasingly shift to creating substantial driving forces motivating the adoption of new technologies. This may allow evolutionary economics to better understand and identify “critical mass” within subsets of the population segmented by social group or social identity.

6.9

Affective psychology

Affective psychology emerged during the neuropsychological revolution of the 1990s, and posed challenges to the strict computational view of the mind invited by cognitive 94

Evolutionary economics and psychology

psychology, instead suggesting the potence and centrality of the visceral emotions. The findings of Joseph Le Doux, for instance, showed that even in the neural structure of visual perception systems, a secondary circuit routed incoming sense-data via the amygdala – the anchor of the brain’s primary fear circuitry – prior to the visual cortex (Le Doux, 1996). A range of other studies surveyed in Jaak Panksepp’s magisterial Affective Neuroscience (Panksepp, 1998) present similar findings demonstrating how the anchors for the primary, base emotional systems are similarly located deep in the brain below the neocortex. This means we “feel” before we even “see” what is causing the feeling. The function of reasoning, which is sometimes called the “higher cortical functions” is not to present a processed view of the world to the emotional systems for reaction, but rather to elaborate the immediate reaction of the emotional systems to the environment. The executive “reasoning” circuitry that we would associate with higher order reasoning and conscious rational decision making, it has been found, serves largely to inhibit or modify the emotional circuitry post hoc (Sapolsky, 2017). The mind is grounded in emotional reactions that provide the human being with the necessary basic motivations to facilitate their survival and reproduction, which are guided and directed by perception and cognitive modelling, and while learning is possible to update the system, it remains anchored in emotional reactions (Asma and Gabriel, 2019). For evolutionary economics, affective psychology offers us an interesting perspective on the adoption of technology that can be important to understanding strong resistance that seems irrational. Emotional reactions can provide as powerful a restraining force as they do a driving force, and if activated can prevent what seems like a fundamental mental “block” on the adoption of technology. Integrating affective psychology into the analytical foundation of evolutionary economics can help us understand the motivation for innovation and entrepreneurship, but it can also help us understand the fundamental barriers that stand in its way.

6.10

Persuasion psychology

It is well known that the neural substrates for cognitive routines decay or strengthen depending on whether they are consistently rejuvenated by their repeated activation (Edelman, 1978). But what is perhaps as interesting for evolutionary economics is the analysis of how new routines can be incorporated into the mind in the first place, which is the subject of persuasion psychology. Persuasion psychology is synthesised in two famous books by Robert Cialdini and the brothers Heath. Cialdini identifies six factors that increase the likelihood that you will get someone to adopt an idea or behaviour that you want them to adopt: your ability to trigger reciprocity, your ability to obtain commitment or demonstrate the consistency of your idea with their existing mindset, your ability to demonstrate that the idea has social proof, your ability to be likeable, your ability to be authoritative, and your ability to create a sense of scarcity (Cialdini, 1984). The brothers Heath similarly identify six factors, albeit different factors that can be summarised by the acronym SUCCES; an idea is more persuasive the more it is Simple, Unexpected, Concrete, Credible, Emotional, and tells a Story (Heath and Heath, 2007). Ultimately, persuasion can be somewhat reduced to the phenomenon of cognitive dissonance. Cognitive dissonance arises when two mutually contradictory ideas are held in mind at once and is an uncomfortable feeling that people will seek to either avoid or reduce to restore consonance (Festinger, 1957). Simply put, once you get someone’s attention, you are 95

Brendan Markey-Towler

more able to persuade someone if you avoid the suggestion of inconsistency between the idea you want them to adopt and the ideas in their mind at the time. For evolutionary economics, persuasion psychology offers us the opportunity to better understand the psychological factors affecting absorptive capacity, which is a vital factor in economic evolution (Cohen and Levinthal, 1990). The dissonance new ideas create a strong restraining force that interacts with absorptive capacity. Integrating persuasion psychology into evolutionary economics therefore can help us understand absorptive capacity for new technologies across the economy at a new level.

6.11

Conclusion: Where we are, where we could go

In this chapter, we discussed the broad sweep of traditions in evolutionary economics and their psychological foundations. We saw that the neo-Schumpeterian, Veblenian, and “Naturalistic”, or “bioeconomic” perspectives and found them to be grounded in and consistent with cognitive and evolutionary psychology, with which they co-developed historically. While this creates a singular advantage for evolutionary economics among other schools of economic thought, there is potential for growth by integrating other fields of psychology. We discussed the opportunities presented by four subfields: personality, social, affective and persuasion psychology. Each presents us with potential to expand our understanding of the driving and restraining forces influencing the tendency for individuals in society to adopt or resist new technologies either in their employment, or in their broader lifestyle. Some work has been done seeking to integrate, for instance, personality and persuasion psychology (particularly cognitive dissonance) into economics more generally by Peter Earl with his ongoing research into lifestyle economics and corporate “imaginations” (Earl, 1984, 1986; Earl and Wakeley, 2010). The personalities that build and underlie lifestyles and corporate imaginations (roughly, operating systems), in this work can be understood to provide significant restraining forces preventing the adoption of new technologies where this would create cognitive dissonance and vice versa where they create consonance. Other work has been done seeking to lay formal groundwork that provides a “bridge” whereby these traditions in psychology may be carried across into evolutionary economics (MarkeyTowler, 2018). But further work that extends on these attempts is possible, and desirable, for evolutionary economics to further integrate psychology proper into its analytical foundations and gain new ground for its explanatory power.

References Asch, Solomon (1956). Social Psychology. Englewood Cliffs: Prentice Hall. Asma, Stephen T. and Gabriel, Rami (2019). The Emotional Mind. Cambridge, Massachusetts: Harvard University Press. Bandura, Albert (1977). Social Learning Theory. Englewood Cliffs: Prentice Hall. Cialdini, Robert (1984). Influence. New York: HarperCollins. Cohen, Wesley M. and Levinthal, Daniel A. (1990). Absorptive capacity: A new perspective on learning and innovation. Administrative Science Quarterly. 35(1): 128–152. Cosmides, Leda and Tooby, John (2013). Evolutionary psychology: New perspectives on cognition and motivation. Annual Review of Psychology. 64:201–229. Cyert, Richard and March, James (1963). A Behavioral Theory of the Firm. Hoboken: Wiley-Blackwell. Digman, John M. (1990). Personality structure: Emergence of the five-factor model. Annual Review of Psychology. 41:417–440.

96

Evolutionary economics and psychology Dopfer, Kurt (2004). The economic agent as rule maker and rule user: Homo Sapiens Oeconomicus. Journal of Evolutionary Economics. 14:177–195. Dopfer, Kurt, Foster, John and Potts, Jason (2004). Micro-meso-macro. Journal of Evolutionary Economics. 14:263–279. Dopfer, Kurt and Potts, Jason (2007). The General Theory of Economic Evolution. London: Routledge. Earl, Peter E. (1984). The Corporate Imagination. Brighton: Harvester Wheatsheaf. Earl, Peter E. (1986). Lifestyle Economics. Brighton: Harvester Wheatsheaf. Earl, Peter E. and Wakeley, Tim (2010). Alternative perspectives on connections in economic systems. Journal of Evolutionary Economics. 20(2):163–183. Edelman, Gerald (1978). Neural Darwinism. New York: Basic Books. Festinger, Leon (1957). A Theory of Cognitive Dissonance. Palo Alto: Stanford University Press. Goldberg, Lewis R. (1993). The structure of phenotypic personality traits. American Psychologist. 48(1):26–34. Heath, Chip and Heath, Dan (2007). Made to Stick. New York: Random House. Hermann-Pillath, Carsten (2013). Foundations of Economic Evolution. Cheltenham: Edward Elgar. Hodgson, Geoffrey (1998). The approach of institutional economics. Journal of Economic Literature. 36(1):166–192. Hodgson, Geoffrey (2004). The Evolution of Institutional Economics. London: Routledge. Hodgson, Geoffrey (2010). Choice, habit and evolution. Journal of Evolutionary Economics. 20:1–18. Hodgson, Geoffrey and Knudsen, Thorbjorn (2010). Darwin’s Conjecture. Chicago: University of Chicago Press. James, William (1890). The Principles of Psychology. Mineola: Dover. Kahneman, Daniel (2011). Thinking Fast and Slow. New York: Farrar, Strauss and Giroux. Kahneman, Daniel (2003). Maps of Bounded Rationality: Psychology for Behavioral Economics. American Economic Review. 93(5):1449–1457. Kelly, George A. (1963). A Theory of Personality. New York: W.W. Norton. Le Doux, Joseph (1996). The Emotional Brain. London: Orion. Lewin, Kurt (1938). The Conceptual Representation and the Measurement of Psychological Forces. Durham: Duke University Press. March, James and Simon, Herbert A. (1958). Organizations. Hoboken: Wiley. Markey-Towler, Brendan (2018). A formal psychological theory for evolutionary economics. Journal of Evolutionary Economics. 28:691–725. Nelson, Richard and Winter, Sidney (1982). An Evolutionary Theory of Economic Change. Cambridge, Massachusetts: Harvard University Press. Newell, Alan (1990). Unified Theories of Cognition. Cambridge, Massachusetts: Harvard University Press. Newell, Alan and Simon, Herbert A. (1972). Human Problem Solving. Englewood Cliffs: Prentice-Hall. Panksepp, Jaak (1998). Affective Neuroscience. Oxford: Oxford University Press. Pinker, Steven (1997). How the Mind Works. New York: W.W. Norton. Sapolsky, Robert (2017). Behave. London: Penguin. Schumpeter, Joseph (1911). The Theory of Economic Development. Piscataway: Transaction. Schumpeter, Joseph (1942). Capitalism, Socialism and Democracy. New York: HarperPerennial. Simon, Herbert A. (1947). Administrative Behavior. New York: Free Press. Tajfel, Henri (1974). Social identity and intergroup behaviour. Social Science Information. 13: 65–93. Thaler, Richard (2015). Misbehaving. New York: W.W. Norton. Veblen, Thorstein (1898). Why is economics not an evolutionary science? Quarterly Journal of Economics. 12(4):373–397. Veblen, Thorstein (1899). The Theory of the Leisure Class. Oxford: Oxford University Press. Veblen, Thorstein (1914). The Instinct of Workmanship and the State of the Industrial Arts. London: Taylor & Francis. Wilson, Edward Osborne (1975). Sociobiology. Harvard: Harvard University Press. Witt, Ulrich (1999). Bioeconomics as economics from a Darwinian perspectives. Journal of Bioeconomics. 1:19–34. Witt, Ulrich (2001). Learning to consume – A theory of wants and the growth of demand. Journal of Evolutionary Economics. 11:23–36.

97

7 EVOLUTIONARY CULTURAL SCIENCE Carsten Herrmann-Pillath

7.1

Introduction

In the recent two decades, evolutionary cultural science (ECS) has been emerging as a transdisciplinary field crossing evolutionary economics, biological theories of coevolution, and the humanities (Hartley and Potts 2014). This development concurs with the revival of culture as a topic in general economics (Beugelsdijk and Maseland 2010; Alesina and Giuliano 2015). In these approaches, culture is mostly treated as an exogenous parameter of behavioural phenomena that can be scrutinized by various methods, most prominently experiments informed by game theory. These parameters are often retrieved from value studies, such as large-scale surveys, or by employing direct measurements of attitudinal stances, such as on authoritarian values. Culture is conceived as being shared among groups of people, mostly ethnic or national groups, and as being transmitted across generations. ECS differs from these economic approaches in important respects, as it is grounded in evolutionary theory, both in a direct sense, such as when explaining preferences in terms of human phylogeny, as in terms of formal homologies, such as employing evolutionary models on the diffusion of consumption patterns. ECS starts out from the thesis of ontological continuity between biological and economic evolution (Witt 2003). ECS touches upon the entire range of economics topics, but in this short entry, we focus on one topic that was prominent in the incipient stage of ECS research: Consumption.

7.2

Ontology and evolutionary methodology

Culture is a highly elusive term with many definitions and uses, which also affects how new approaches such as ECS locate themselves in this vast field. There are the following two distinct approaches. • The first is represented in the rise of co-evolutionary theory in biology (seminally, Boyd and Richerson 1985). Here, culture is a generic capacity shared by all members of the human species. This results in broad definitions of culture, referring to all kinds of biological information that are transmitted cross-generationally in non-genetic media (Mesoudi 2011).

98

DOI: 10.4324/9780429398971-9

Evolutionary cultural science

Such broad definitions are also prominent in anthropology, where culture often encompasses the entire domain of the ‘artificial’ that defines the diversity of human groups, such as religious beliefs, food practices, or all kinds of artefacts (Brumann 1999). Combining the two views, we arrive at the notion of coevolutionary niche construction (Odling-Smee et al. 2003). The genetically endowed human capacity of culture enables phenotypic plasticity far beyond what is genetically determined, such as adapting to vastly different local climates and ecological conditions. • However, anthropology also features the second view on culture which is dominant in the humanities, namely culture as a manifestation of human diversity freed from biological fetters (Sahlins 1976). This concurs with a similar duality in the study of language: There is a generic human capacity for language, yet humans do not speak one language but many tongues which are mutually unintelligible, though share some basic properties. Interest in human diversity also prevails in economic studies of group effects on behaviour, such as ethnic differences. The first view also accounts for human diversity since the notion of adaptive niche implies cultural diversity, though following certain functional requirements of adaptation. This view was most radically defended by sociobiology in its heyday (Wilson 1975), but also by materialist approaches in anthropology (Harris 1979). In this view, cultural differences can be explained by causal mechanisms that relate observed behaviour to certain external conditions, such as explaining food habits as adapting to local ecology. This is one version of co-evolutionary theory: Culture is not genetically transmitted but obeys to the same principles as biological evolution. Hence, a major research concern is the mechanisms that explain the transmission and retention of cultural traits and identify the adaptive functions of those traits. This establishes a formal homology between biological theories of evolution and cultural science that frames direct mechanisms of gene-culture coevolution (Richerson et al. 2010; Altman and Mesoudi 2019). However, human cultural diversity has also motivated the opposite conclusion, namely that culture is the form of life that constitutes human autonomy from biology. This was also a key idea in constituting the autonomy of the humanities as a scientific endeavour, as ‘Geisteswissenschaft’ or ‘Kulturwissenschaft’, which relates to the distinction between two ways of doing research, ‘Erklären’ and ‘Verstehen’ (Dilthey 1883). This distinction is also salient in economic debates over culture, which today focus on Erklären: In contrast, Verstehen as cultural hermeneutics is a key concern in the so-called ‘old institutionalist’ approach to evolutionary economics informed by American pragmatism, which is still prominent among its followers, but also, for example, in Austrian economics research (Grube and Storr 2015). In terms of evolutionary approaches, the distinction between Erklären and Verstehen can be referred to as the more fundamental distinction between atomistic and holistic or systemic ontologies (Bunge 1977). This is also manifest in modern biology. The focus on traits and genes in the Neodarwinian synthesis grounds in an atomistic ontology combined with the standard model of causal explanation, often resulting in explanatory reductionism. However, this view has met considerable criticism in recent decades, with systemic approaches that unify evolution and development in one conceptual frame and theories of multi-level evolution that result in a pluralist view of biological information and its transmission (Baedke and Gilbert 2021). To a large extent, this view comes close to the Verstehen pole of the scientific method (Gould 2002). 99

Carsten Herrmann-Pillath

In the holistic view, there is no direct causal linkage between external conditions and traits, since traits are always embedded into systems that generate adaptive performance on the systems level. The function of the trait is its meaning in the system (Wuketits 2005). For example, a plant is embedded into a local ecosystem, which ties up with larger regional systems, and so on. Since the lower-level environment is endogenous to these larger systems, ultimately function is deflated to general criteria such as reproducibility, sustainability, and resilience. This is formally homologous to the impossibility of assigning meanings to words of a language only based on direct causal chains that connect external events with the usage: The meaning or function as usage is defined in the entire context of meanings as constituted by other words of the language (Quine 1960). Hence, Verstehen requires the reconstruction of this systemic whole, a key concern of ‘cultural science’ as Geisteswissenschaft. Further, the holistic ontology approaches genuine individuality as a correlate of systemicity, constituted on the highest level, that is, treating systems as genuine individuals, and consequently, all constituent parts (Herrmann-Pillath 2013). Formally, this corresponds to the role of randomness in the Neodarwinian theories of evolution and is crucial for explaining the creative powers of evolution: Genuine novelty grounds in genuine individuality, which is treated as a generic phenomenon, hence randomness, in the populationcentred explanatory models. In contrast, the Verstehen view demands specification of individuality, such as in historical explanations of social change that would centre on the role of single agents, specific localities, and so on. That means ‘individuality’ is not understood as another manifestation of atomistic ontology but refers to pervasive contextualization of all elements of a causal trajectory of change. Thus, eventually, the dualism of Erklären and Verstehen is overcome by the notion of contextualization of causal explanation. In economics, one important example is the role of entrepreneurship in generating novelty: For explaining technological change, we must understand structural determinants and mechanisms of diffusion, but eventually entrepreneurial individuality looms large, and how this interacts with idiosyncratic factors on all levels (the ‘Schumpeterian entrepreneur’ of evolutionary economics). Hence, evolutionary analysis of economic change is deeply informed by historical perspectives in the sense of recognizing the individuality and contextuality of all trajectories of change: This is the notion of the historicity of evolution (Dopfer 2005).

7.3

Principles of ECS

ECS can now be defined in more specific terms. First, the term ‘science’ implies that ECS is basically pursuing causal explanations of observed phenomena in human ways of life, while overcoming the false dichotomy between Verstehen and Erklären, on both sides. A corollary of this is second, that ECS closely associates with economics, which, in the larger domain of the humanities and the social sciences, is perhaps the most developed discipline in terms of pursuing causal explanations. Yet, again following from the previous, ECS is not compatible with those economic approaches that build on axiomatic frameworks that deny genuine individuality and pursue complete de-contextualization, such as, for example, in many macroeconomic theories. Economic allies of ECS include programmatic alternatives, most importantly evolutionary economics, but also behavioural economics, institutional economics, or, after all, cultural economics. The role of economics also reflects another characteristic, namely the focus on materiality. 100

Evolutionary cultural science

Materiality has become a key concern in many strands of thought in the humanities and the social sciences. One of the leading areas is economic sociology (Pinch and Swedberg 2008). Further, materiality has been emerging in other fields that connect with economics, most significantly, cognitive sciences, which explore the role of distributed forms of cognition (Clark 2011). Hence, materiality means that ECS eschews internalist and abstract models of thought, choice, and action in favour of externalist approaches that highlight the role of external constituents (Ross 2014). This view directly matches the systemic views on context in causal explanations. The third aspect of defining ESC is the notion of evolution. This is important in several respects. Most fundamentally, culture is about change. This needs emphasis as many uses of culture in economics follow static views of culture that treat culture as a (exogenous) constant, reflecting the idea that culture is transmitted from the past to the present (Jones 2006). This motivates false essentialization of culture, such as treating national cultures as givens (Hofstede et al. 2010). This is empirically misleading, as culture manifests wide variety across groups, is mostly constituted by cross-cultural interactions, and is expressed in creative action, though joined with habit. This creative dimension of culture has been much emphasized by the ‘old’ institutional evolutionary economists, informed by pragmatism. Evolutionary economics has a rich discussion about various approaches to evolution, especially regarding the question of whether biological theories apply to economic analysis (Nelson 2006; Hodgson and Knudsen 2013). As we have seen, the notion of evolution is also open to various interpretations in biology. ECS follows the biological debate in this wider sense. Accordingly, there are narrow and broad interpretations of evolution. The narrow interpretations follow the Neodarwinian synthesis in transferring the rich modelling resources to the study of culture. Memetics is one example, but also more generally all kinds of research on the diffusion of cultural traits in human populations (Schlaile, this volume; Mesoudi 2011). The broad interpretation emphasizes systemicity and context and relates to the study of phylogeny, taxonomy, and, in general terms, the approach of the naturalist to culture, which comes close to classical field anthropology (Gould 2002). In considering the difference between the narrow and the broad views on evolution, perhaps the key issue is the question of agency in cultural selection. In Neodarwinian models, agency is either completely diluted in the population-level processes or recognized in positing various generic mechanisms of learning (Mesoudi 2011). In all these approaches traits remain given, such as copying a certain behaviour from others. In contrast, the broad conceptions allow for creative agency. That means traits are interpreted by individual agents in an autonomous way; hence, giving meaning to them. This leads to considering the relationship between function and meaning in cultural selection (Rogers and Ehrlich 2008). A trait that directly meets a functional requirement, such as using a knife, may diffuse rapidly in a population, with everyone noticing the functional expedience of the tool. In contrast, most expressive forms of behaviour have much interpretive leeway, such as ways of dressing, music, the arts, or religion. Hence, much cultural diversity reigns in the expressive domain. Active interpretive agency roots in reflexivity, absent in mechanistic models of selection (seminally, Mead 1932). Reflexivity introduces a cognitive filter between external causes and behavioural responses. This is most important when it comes to interactions between humans, as reflexivity requires reconstructing Alter’s internal cognitive operations, hence the intentions, resulting in noncomputable coordination problems (Zawidzki 2013). Evolutionary theory thus assigns a specific function to culture in the expressive domain, namely resolving issues in behavioural 101

Carsten Herrmann-Pillath

coordination by the cognitive shortcut of external anchoring of mutual perceptions and expectations. This suggests a definition of culture that is much narrower than most other generic ones and which has also been suggested by economists (Beugelsdijk and Maseland 2010). Culture is no longer equated with everything non-natural, but with a specific function, namely mediating the formation of stable identities of actors in social interactions with reflexivity. Culture is the medium of ‘mind shaping’ as a fundamental human activity, such as teaching others to do things properly and being receptive to such guidance (Zawidzki 2013). Accordingly, economic phenomena such as consumption cannot be adequately understood by focusing on individual choice, as in the standard economic model.

7.4 ECS behavioural theory of consumption ECS has many implications for economics. One important field is the study of consumer behaviour, which closely ties up with identity when considering the expressive function of consumption. The evolutionary theory of consumption departs fundamentally from the canonical economic model. There are three main determinants of consumption: • The coevolutionary grounding of preferences. • The formation and expression of identity. • The role of status and social network effects. ECS distinguishes between human needs and wants, approaching the former as biologically grounded and the latter as culturally evolving (Witt 2003). There is a wide range of human needs which manifest species-specific characteristics and phenotypic plasticity. Many consumption patterns can be explained as adaptive functions, such as wearing a warm coat in cold climate. In combination with technological progress, such functional needs multiply, such as the need for cooling in storage. Accordingly, concepts such as basic needs can be used to assess the level of economic development and human well-being (Corning 2003). At the same time, classical models of preference evolution such as Maslow’s hierarchy of needs suggest that with growing affluence, needs become decoupled from biological factors, narrowly defined. However, for ECS, one of the key biological needs is the need for recognition by others and the aspiration to secure relative status in a group (Saad 2007). This is a feature shared with primates living in groups, as relative status is a prime determinant of access to mates and resources, while containing escalation of violent conflicts. Further, cooperation in groups and competition among groups is a defining element of human niche construction (Bowles and Gintis 2011). ECS suggests a general model of evolving preferences over status and positional goods, which drives differentiation of consumption patterns independent of the level of economic development. Status good consumption implies that there is no satiation point as differentiation can be always increased and ego’s satisfaction is always co-determined by the position relative to others (classically, Veblen 1899; Frank 1985). This has direct effects on socio-economic evolution of consumption patterns, since the distinction between basic goods and status goods as ‘luxury’ results in moving the ‘consumption frontier’ forward, a mechanism already highlighted by Adam Smith. Examples abound, such as the diffusion of tea and sugar in England, both being non-existent in the native consumption patterns and 102

Evolutionary cultural science

eventually turning from luxury to everyday consumer items (Trentmann 2017). In general, the ‘industrious revolution’ in Western Europe was accompanied, if not driven, by the formation of a material consumer culture emphasizing a wide range of household possessions as markers of status (De Vries 2008). Such effects continue driving the emergence of material consumer culture in catching up economies such as China. Status can be conceived in narrow and broad senses. The narrow sense is about relative position in a social status order. The broad sense is status as individual identity in the context of the group, that is, social identity (Akerlof and Kranton 2000). Postmodern societies manifest much leeway in ranking and dimensionality of status, such that this broader conception of status prevails, as in the phenomenon of lifestyle consumption. Consider food consumption. The need for nutrition appears to be a biological function. But humans learn about the appropriate food while growing up in human groups living in specific places, and they have only a limited inborn capacity to identify the appropriate food. Therefore, independent from level of development, human groups have evolved rich and locally circumscribed food cultures. Often, these food cultures are key elements in defining the identity of groups, such as food taboos. In postmodern societies, globalization has resulted in highly diverse and pluralist food practices. Yet, the many phenomena of eating disorders and serious health hazards from dysfunctional eating habits reveal a gap, if not even a contradiction, between wants and needs (Berridge 2009). Often, these phenomena are caused by identity issues, such as self-perceptions of beauty, strongly influenced by peer group effects. ECS distinguishes between social identity and personal identity (Davis 2010). Personal identity is embodied in biographical memory and in reflexive processes which allow for considering choices ex ante and ex post, drawing conclusions from them, and eventually being able to give reasons, both in inner speech and in conversation with others. In contrast, social identity is bound to the recognition of others. The group dynamic is shaped by the two factors of positive recognition and negative sanctions, the latter being rooted in a fundamental biological drive to moral aggression, that is, discrimination and ostracism directed at deviant group members or members of other groups (Bowles and Gintis 2011). The human penchant for discriminating ingroup and outgroup, though deplorable, is a behavioural universal. The evolutionary explanation is straightforward, as ostracism is a powerful means to sustain group-bound cooperation and altruism, which may be individually costly in the narrow scope, yet contributes to group survival and hence also individual benefit (Sober and Wilson 1998). Hence, adopting consumption patterns in groups is partly tied to the costly signalling of social identity, since cooperation can be undermined by cheating and free-riding. Costly signalling is a marker of group identity which cannot be easily imitated or faked (Zahavi and Zahavi 1997). If we move beyond the original phylogenetic setting, these phenomena persist in modern societies and provide another explanation for the differentiation of status goods. Social network effects are especially strong for those goods which do not meet a direct functional need. The obvious case is the entire range of cultural goods and industries, including fashion and lifestyle products. These are ‘social network markets’, where the consumption is not grounded in individual preferences, but social preferences (Potts et al. 2008). That means peer group effects and imitation drive individual choices or personal needs to distinguish oneself from the crowd. However, there are also less obvious domains, most significantly, finance, where network effects loom large (Preda and Gemayel 2020). 103

Carsten Herrmann-Pillath

To sum up, ECS posits that individual preferences and material cultures co-evolve under a regime of competition for status within and between groups. By implication, we cannot define a de-contextualized benchmark for assessing ‘progress’ in meeting given human needs, which transpires in the complex picture of mapping economic measures of growth with indicators of ‘happiness’ as subjective well-being (Helliwell et al. 2021).

7.5 Outlook: The culture of economics ECS encompasses all domains of economic analysis since culture is a key contextual determinant of economic processes. Beyond consumption, other important fields are economic growth and development, technological change, and institutions. The role of culture in comparative economic development is widely recognized, both in the longue durée and with reference to shorter timespans of development policies. In the very long run, culture appears as crystallizing various complex factors of technology, institutions, and ecology, such as when considering the ‘Great divergence’ between Imperial China and Europe during the past millennium. Culture remains a key notion in the current geopolitical setting of systems competition between emerging China and the old hegemonial powers (Herrmann-Pillath 2017). Cultural conflict, in the form of ethnic and religious tension, fractures many countries in subSaharan Africa and undermines the stability of political and economic institutions. On a deeper level even, ECS suggests approaching economic systems as manifesting cultural evolution, contra the widespread belief in economics that the emergence of modern economic institutions reflects functional efficiency and adaptive function. The modern culture of consumption is a major element of capitalism as it emerged in Western Europe and diffused across the globe. Facing the challenge of climate change and the looming catastrophe for human civilization and the biosphere, ECS allies with ecological economics in emphasizing the urgent need to develop new co-evolutionary concepts of economic systems (Goddard et al. 2019). Apart from changing material consumer culture, this requires fundamental cultural transformations. At the core, this means moving from an anthropocentric conception of economics to a geocentric view, thus revealing that economics itself is an essential building block of capitalism (Polanyi 1944). Economic thought is imbued by culture, such as in notions of individualism, property, and welfare, which are amenable to cultural change, thus revealing the performative role of economics as a cultural activity in capitalism. In the same vein, technological evolution has been driven by cultural conceptions, if not hybris, of human mastery over nature. Therefore, a major concern of ECS is developing approaches to the new performativity of economics.

References Akerlof, G. A. and R. E. Kranton (2000): Economics and Identity. Quarterly Journal of Economics, CXV(3), 715–753. Alesina, A. and P. P. Giuliano (2015): Culture and Institutions. Journal of Economic Literature, 53(4), 898–944. 10.1257/jel.53.4.898. Altman, A. and A. Mesoudi (2019): Understanding Agriculture within the Frameworks of Cumulative Cultural Evolution, Gene-Culture Co-Evolution, and Cultural Niche Construction. Human Ecology, 47, 483–497. Baedke, J. and S. F. Gilbert (2021): Evolution and Development, The Stanford Encyclopedia of Philosophy (Fall 2021 Edition), E. N. Zalta (ed.), URL = < https://plato.stanford.edu/archives/ fall2021/entries/evolution-development/>.

104

Evolutionary cultural science Berridge, K. C. (2009): ‘Liking’ and ‘Wanting’ Food Rewards: Brain Substrates and Roles in Eating Disorders. Physiology & Behavior, 97, 537–550. 10.1016/j.physbeh.2009.02.044. Beugelsdijk, S. and R. Maseland (2010): Culture in Economics. History, Methodological Reflections, and Contemporary Applications. Cambridge: Cambridge University Press. Bowles, S. and H. Gintis (2011): A Cooperative Species: Human Reciprocity and Its Evolution. Princeton, NJ: Princeton University Press. Boyd, R. and P. J. Richerson (1985): Culture and the Evolutionary Process. Chicago, IL, and London: University of Chicago Press. Brumann, C. (1999): Writing for Culture: Why a Successful Concept Should not be Discarded. Current Anthropology, 40, Supplement, S1–S28. 10.1086/200058. Bunge, M. (1977): Treatise on Basic Philosophy, Volume 3. Ontology I: The Furniture of the World. Dordrecht: Reidel. Clark, A. (2011): Supersizing the Mind: Embodiment, Action, and Cognitive Extension. Oxford: Oxford University Press. Corning, P. A. (2003): Nature’s Magic. Synergy in Evolution and the Fate of Humankind. Cambridge: Cambridge University Press. Davis, J. B. (2010): Individuals and Identity in Economics. Cambridge: Cambridge University Press. De Vries, J. (2008): The Industrious Revolution. Consumer Behavior and the House¬hold Economy, 1650 to the Present. Cambridge. Dilthey, W. (1883): Einleitung in die Geisteswissenschaften. Versuch einer Grundlegung für das Studium der Gesellschaft und Geschichte. Leipzig: Duncker $ Humblot, http://www.deutschestextarchiv.de/ book/show/dilthey_geisteswissenschaften_1883 Dopfer, K. (2005): Evolutionary Economics: A Theoretical Framework, in Dopfer, K. (ed.), The Evolutionary Foundations of Economics. Cambridge: Cambridge University Press, pp. 3–56. Frank, R. H. (1985): Choosing the Right Pond: Human Behavior and the Quest for Status. New York, NY, and Oxford: Oxford University Press. Goddard, J. J., G. Kallis and R. B. Norgaard (2019): Keeping Multiple Antennae up: Coevolutionary Foundations for Methodological Pluralism. Ecological Economics, 165, 106420. 10.1016/ j.ecolecon.2019.106420. Gould, S. J. (2002): The Structure of Evolutionary Theory. Cambridge, MA, and London: Belknap. Grube, L. E. and V. H. Storr, eds. (2015): Culture and Economic Action. Cheltenham and Northampton, MA: Edward Elgar. Harris, M. (1979): Cultural Materialism: The Struggle for a Science of Culture. New York, NY: Random House. Hartley, J. and J. Potts (2014): Cultural Science: A Natural History of Stories, Demes, Knowledge and Innovation. London and New York, NY: Bloomsbury Academic. Helliwell, J. et al. (eds.) (2021): World Happiness Report 2021, https://happiness-report.s3. amazonaws.com/2021/WHR+21.pdf Herrmann-Pillath, C. (2013): Foundations of Economic Evolution. A Treatise on the Natural Philosophy of Economics. Elgar. Herrmann-Pillath, C. (2017): China’s Economic Culture. The Ritual Order of State and Markets. Routledge. Hodgson, G. and T. Knudsen (2013): Darwin’s Conjecture. The Search for General Principles of Social and Economic Evolution. Chicago, IL: University of Chicago Press. Hofstede, G., G. J. Hofstede and M. Minkov (2010): Cultures and Organizations. Software of the Mind. New York, NY: McGraw Hill. Jones, E. L. (2006): Cultures Merging: An Historical and Economic Critique of Culture. Princeton, NJ: Princeton University Press. Mead, G. H. (1932): Mind, Self, and Society. The Definitive Edition. Chicago, IL, and London: University of Chicago Press, 2015. Mesoudi, A. (2011): Cultural Evolution: How Darwinian Theory Can Explain Human Culture and Synthesize the Social Sciences. Chicago, IL, and London: University of Chicago Press. Nelson, R. R. (2006): Evolutionary Social Science and Universal Darwinism. Journal of Evolutionary Economics, 16(5), 491–510. 10.1007/s00191-006-0025-5. Odling-Smee, F. J., K. N. Laland and M. W. Feldman (2003): Niche Construction: The Neglected Process in Evolution. Princeton, NJ: Princeton University Press.

105

Carsten Herrmann-Pillath Pinch, T. and R. Swedberg (eds.) (2008): Living in a Material World: Economic Sociology Meets Science and Technology Studies. Cambridge, MA: MIT Press. Polanyi, K. (1944/2001): The Great Transformation: The Political and Economic Origins of Our Time. Boston. Potts, J., S. Cunningham, J. Hartley and P. Ormerod (2008): Social Network Markets: A New Definition of the Creative Industries. Journal of Cultural Economics, 32(3), 167–185. 10.1007/s10824-008-9066-y. Preda, A. and R. Gemayel (2020): Scopic Systems and Decision Making in Financial Markets, in Harbecke, J. and C. Herrmann-Pillath (eds.), Social Neuroeconomics: Mechanistic Integration of the Neurosciences and the Social Sciences. London: Routledge, pp. 262–274. Quine, W. V. (1960/2013): Word and Object. New ed. Cambridge, MA: MIT Press. Richerson, P. J., R. Boyd and J. Henrich (2010): Gene-Culture Coevolution in the Age of Genomics. Proceedings of the National Academy of Sciences, 107(suppl. 2), 8985–8992. Rogers, D. S. and P. R. Ehrlich (2008): Natural Selection and Cultural Rates of Change. Proceedings of the National Academy of Sciences, 105(9), 3416–3420. 10.1073/pnas.0711802105. Ross, D. (2014): Philosophy of Economics. New York, NY: Palgrave MacMillan. Saad, G. (2007): The Evolutionary Bases of Consumption. New York, NY: Lawrence Erlbaum. Sahlins, M. (1976): Culture and Practical Reason. Chicago, IL: University of Chicago Press. Sober, E. and D. S. Wilson (1998): Unto Others: The Evolution and Psychology of Un–selfish Behavior. Cambridge, MA, and London: Harvard University Press. Trentmann, F. (2017): Empire of Things: How We Became a World of Consumers, from the Fifteenth Century to the Twenty-First. Penguin History. London: Penguin Books. Veblen, T. (1899/1965): The Theory of the Leisure Class. New York, NY: Kelley. Wilson, E. O. (1975/2000): Sociobiology: The New Synthesis. 25th-anniversary ed. Cambridge, MA: Belknap Press of Harvard University Press. Witt, U. (2003): The Evolving Economy. Essays on the Evolutionary Approach to Economics. Cheltenham and Northampton, MA: Edward Elgar. Wuketits, F. M. (2005): The Theory of Biological Evolution: Historical and Philosophical Aspects, in F. M. Wuketits and F. J. Ayala (eds.), Handbook of Evolution. Weinheim: Wiley-VCH Verlag, pp. 57–85. 10. 1002/9783527619719.ch4. Zahavi, A. and A. Zahavi (1997): The Handicap Principle: A Missing Piece of Darwin’s Puzzle. New York, NY, and Oxford: Oxford University Press. Zawidzki, T. (2013): Mindshaping: A New Framework for Understanding Human Social Cognition. Cambridge, MA: MIT Press.

106

8 EVOLUTIONARY ECONOMICS AND ECONOMIC HISTORY Andreas Resch

The relations between economic sciences and economic history (as a branch of science) usually are manifested in the following three ways. 1) Application of methods of mainstream economics on historical data in the manner of econometric/cliometric studies. This is the most common approach. 2) Application of heterodox economic theories to research in economic history. The neoinstitutionalist-oriented economic history can be considered the greatest success of this scientific direction, for which, for example, Douglass North was awarded the Nobel Memorial Prize in economics. 3) A further common practice is represented by research in economic history conducted in the historical tradition, which is only loosely influenced by mainstream economics. In this chapter, another particularly promising alternative is to be considered, namely a close connection between evolutionary economics and economic history. The following two chapters serve as a starting base. First, the ontological and heuristic foundations of evolutionary economics are contrasted with those of mainstream economics; thereafter, major strands of historical research since the 18th century are elaborated. Finally, common research perspectives for an evolutionary economic history are discussed.

8.1

Ontological and heuristic foundations of evolutionary economics and mainstream economics

Evolutionary economics1 differs from mainstream economics more than most other heterodox schools in terms of ontological and heuristic foundations. Mainstream economics emerged in the late 19th century in the sense of a ‘hard’ science, under the influence of classical mechanics, the assumption of invariant laws and thus oriented towards deterministic models. The central concepts used to capture economic phenomena include equilibrium, homogeneity of goods, capital and information, representative firms and households, given technologies and preferences. The central heuristic is the concept of thinking about economics from the perspective of homo oeconomicus, the fully informed, rationally optimizing actor under given constraints. A macroeconomic welfare-optimal equilibrium results from individualistic action. Homogeneity and additivity allow simple aggregation from the micro to the macro level. DOI: 10.4324/9780429398971-10

107

Andreas Resch

Theory construction is used instrumentalistically for the organization of the methodical research process. This also allows, for example, to see the rationality assumption merely as a methodological concept whose usefulness can be discussed, but which cannot be disproved. Everything that cannot be grasped on the basis of the stated ontology and methodology does not belong to its own scientific domain. Change and history act externally and do not occur from within the domain under study. An important quality of this kind of definition of the scientific field is that the demarcation from related fields is very clear. As an example of this conclusive demarcation, see for instance how Gul and Pesendorfer (2005) delimit canonical economics from neuroeconomics. They argue that neuroeconomics, based on physiology and psychology, uses different methods to address different questions than mainstream economics. Therefore, it cannot refute any economic models or findings. Consequently, they argue, it makes no sense for economics to expand into the other field of science. In comparison to canonical economics, evolutionary economics is a patchwork of approaches, questions, and methods. It constructs its research field in a more interdisciplinary way as a complex, open system. Consequently, it is not based on a unified master theory. Common conceptual foundations are the complexity assumption, endogenous emergence of novelty, dynamics of variation, diffusion, selection, retention, and self-organization processes. Evolutionary economics is not confined to post-disturbance processes, which means that it deals with processes, which (partially) transcend the possibilities of formal modeling. Among the basic ontological assumptions is the heterogeneity of actors, technologies, firms, information, etc. Due to endogenous emergent change, developments proceed as open processes. Information and knowledge change imply the central importance of novelty and learning. In view of complexity, rationality can at best be replaced by ‘bounded rationality’. Developments are characterized by individual events (singularities) and actors as well as by complementarities, paths or ‘trajectories’, non-linearities, non-ergodicity, contingency, and (social) structures. Instead of stable equilibria, at best meta-stable states can be achieved, including inefficient lock-in situations. Actors act according to routines and cannot optimize, but strive for ‘satisficing solutions’. Economic development takes place in the cultural domain and it is agreed that there are essential differences between biological and cultural evolution. While biological evolution proceeds slowly and randomly over generations by way of genetic variation, selection, and retention, cultural evolution can progress much faster thanks to purpose orientation, learning ability, experience, convictions, conscious decisions, reflective ability, emerging institutions, etc. As a consequence of non-linearities, complementarities, and the emergence of qualitatively new phenomena, in an evolutionary setting, the micro level cannot simply be aggregated additively to the macro level (Peneder 2017). Research strategies and theory building focus on the micro level, the macro level, and the meso level, which is necessary for understanding complex interaction processes. An overall conception of this, in which individuals act in a socially organized system, stems from Kurt Dopfer and Jason Potts. They explain dynamics through the emergence of new ‘rules’ in micro- and meso-trajectories (Dopfer, Potts 2004, 2008, Dopfer 2016). Actors are carriers of rules and actualizations of rules. Because of this dualism, diversity rather than uniformity is present. The interdependent developments on the diverse levels can also be structured differently in the time dimension and may take place in complex sequences. Literature surveys reveal that neo-Schumpeterian approaches are at the center of evolutionary economics research. In the tradition of Nelson, Winter (1982), developments of 108

Evolutionary economics and economic history

routines in firms, technological development in innovation systems, innovation competition, industrial dynamics, and resulting long-term development are studied (Nelson et al. 2018). Related to this are topics such as habits and routines in learning processes, self-organization processes, the nature of innovation, demand and cumulative causation of technological change, and increasing returns to learning (Robert et al. 2017). This literature is limited to market capitalist systems. Around this core, further areas of evolutionary economics can be identified. Ulrich Witt (2008, 2013) considers besides neo-Schumpeterian research especially what he calls ‘naturalistic’ approaches to be relevant. These deal with long-term issues such as institutional development, production, consumption, growth, and sustainability. They consider cultural evolution together with the biological evolution of humans. Witt names Veblen, GeorgescuRoegen, Hayek, and North as important sources of inspiration for this research direction. Furthermore, Witt and Andreas Chai (2019) consider contributions in the wake of the game-theoretic revolution and the rise of experimental economics as a new wave of evolutionary economic research. Beyond this, relevant contributions to aspects of evolutionary economics also come from the broader field of research concerned with cultural and political evolution. Seminal publications such as Bowles, Gintis (2011), Boyd, Richerson (1985), Henrich (2016), Mesoudi (2011), and Richerson, Christiansen (2013) come to mind. Nathan Nunn (2020) for instance provides an overview of this kind of cultural evolution research. The work referenced therein is largely done separately from the aforementioned streams of evolutionary economics. The bibliographies in the surveys of Nelson et al. (2018) and Robert et al. (2017), which focus on neo-Schumpeterian research, show little correspondence with that of Nunn (2020). One of the few ‘connecting’ scholars is Joel Mokyr. He links long-term cultural evolution to the ‘central core’ of evolutionary economics with his approach to the evolution of ‘useful knowledge’ (Mokyr 2002, 2018). Cordes and Witt also recommend a closer intertwining of economic and cultural evolution (Cordes 2019, Witt 2013). In addition to the question of how narrowly evolutionary economics should be thematically delimited, various different temporal horizons and degrees of generality of intended statements are also discussed. Representatives of the aforementioned ‘naturalistic’ perspective point out that cultural evolution should be seen together with biological evolution. To study biological and cultural coevolution would require the consideration of extremely long periods of time, since the biological evolution of mankind has taken place over at least several hundred thousand years. Witt links the acceptance or non-acceptance of the continuity hypothesis to the distinction between proximate versus ultimate explanations. He places the former, how something works, in short-term, operational contexts, while for ultimate explanations of why-questions, he considers the study of long-term developments necessary (Witt 2008, Witt 2013). Dopfer makes the important distinction between, on the one hand, the ‘operational level’, at which operations take place at the given level of knowledge, and the ‘knowledge level’, at which the evolutionary development of the knowledge structure is taken into account. In this sense, the shortest time unit or sequence of the investigation of an evolutionary process is a micro- and/or a meso-trajectory, i.e. the process from the emergence of a new idea/rule to its retention (Dopfer 2011). In an overview, Witt and Chai distinguish between addressing historically unique events, analyses of common classes of events, and generalized explanations for evolutionary economic change. They classify singularities as of little interest to evolutionary economics, state 109

Andreas Resch

that neo-Schumpeterian research on industrial transformation belongs to the second form, and recommend looking at the motivational properties of actors as a possible general/ ultimate explanatory basis (Witt, Chai 2019). However, this in turn has to be explained as a result of extremely long-term biological-cultural processes and current cultural processes, which makes it clear that argumentative horizons for aiming at ‘ultimate’ explanations are hardly clearly delimitable due to the problem of infinite regression of ‘why’s’ that proximate causes create. The survey has shown that the field of mainstream economics is logically clearly determinable, while the edges of evolutionary economics are less precisely defined in terms of interdisciplinary openness. As a central difference, it is to be emphasized that newness and change are endogenous and are the main topic in evolutionary economics, whereas in neoclassics they are regarded as exogenous from the ontological foundations. Yet, it should be pointed out here that a picture of mainstream economics as a purely static science limited to perfect markets and fully informed optimizing agents would only be a caricature of actual research practice. In reality, even economists working from the canonical disciplinary core address issues such as growth, technical change, imperfect markets, asymmetric information, bounded rationality, disequilibria, etc. (Coyle 2018, Gul, Pesendorfer 2005). Many authors also take at least implicitly evolutionary positions (Nelson et al. 2018). Nevertheless, mainstream economics will tend to be focused on the operational level, whereas evolutionary economics primarily deals with the knowledge level in the sense of Dopfer (2011).

8.2

History of history

History deals with complex, human-made developments over time, economic history deals with complex human-made economic developments.2 In the European tradition, until the early-modern period, history primarily served to provide theology, philosophy, and jurisprudence with didactically illustrative stories. In the 18th century, it began to establish itself as an independent science concerned with the development of humankind. This also led to its separation from natural history. The new concepts of dynamically developing nature in this field created the intellectual preconditions for the evolutionary biological theories of the 19th century. Human history was now understood as a dynamic, open process resulting in complex ways from the actions of diverse actors in their historical environment. For the humanhistorical domain, one explicitly rejected monocausal explanations and mechanistic concepts of universally valid laws, which emerged simultaneously in the early-modern era for the natural sciences. Historical research was not to be conducted deductively, but primarily empirically and inductively, starting from written sources. The time before mankind developed writing was assigned to a separate subject referred to as prehistory. Around 1800, the idealistic philosophy of history developed general, Eurocentric concepts of historical dynamics of peoples, nations, or humanity. Against this background, also Marxist historical materialism emerged around the middle of the 19th century. In the German university system, which experienced a rapid boom after the Napoleonic Wars, history was able to establish itself as a powerful, independent scientific discipline. Historiographical research was concerned with collective entities such as states, nations, or peoples and the historic impact of ‘great men’ in the field of tension between historiographical developmental thinking and the assumption of diverse, individual actors. It experienced its heyday in Germany in the period around the founding of the German Reich (1870/1). In the 110

Evolutionary economics and economic history

late 19th century, not only in Germany, historiography became associated with doctrines such as biologism and social Darwinism (Hawkins 1997), whereby it regressed behind the separation from natural history that had already taken place in the 18th century and produced ideologies that legitimized imperialism, racism, and radical nationalism. In this intellectual environment, during the first half of the 19th century, the German development theorist Friedrich List formulated his concepts of active national economic development, and later the historical school of national economics emerged, which was opposed to the nascent neoclassical economics and the early Austrian school in the socalled ‘Methodenstreit’. Historical economists such as Gustav von Schmoller relied on historical-empirical research that examined sociologically oriented development, while theoretical economists started from timeless concepts such as that of homo oeconomicus. Joseph A. Schumpeter bridged this conflict in the early 20th century by expressing appreciation for both directions; one for the field of statics, the other for development issues (Andersen 2006, 10–14). Critical approaches to the further advancement of historical science gained momentum in the 20th century in the face of two world wars and the rise of totalitarian systems. The main currents were on the one hand the further development of the humanities in confrontation with naturalism, positivism, and mechanistic concepts, which bet on the further refinement of hermeneutic methods. On the other hand, flourished approaches that systematically combined research from the perspective of diverse actors with that of overall developmental structures (upward and downward causality) using structural approaches in the broadest sense. These intended to make scientific approaches such as sociology, geography, demography, psychology, and, not least, economics usable for historical science. For Europe, until the 1970s, the ‘Annales School’ in France and the ‘Social History as Historical Social Science’ in Germany should be mentioned, both of which had a very strong focus on economic history. Totalitarian ideologies, which claimed deterministic-mechanistic concepts of history, inspired Karl Popper to his critique of what he called ‘historicism’, which thus contributed to the understanding of historical development as an open process, which yet was not new for the historical guild anyway (Popper 2002). Postmodern approaches, cultural and linguistic turns, gender history, global history, etc., have criticized older approaches, contributed to even more diversity, and, in particular, have again emphasized the subjective, individual side of history more than structural considerations. Similar to evolutionary economics, historical studies are such a large and diverse field of research that it cannot be based on a unified theoretical foundation and methodology.

8.3 Mutual considerations: Evolutionary economics as economic history According to the definitions given above, economic history generally deals with complex, human-made economic developments, evolutionary economics with evolutionary processes of economic development in time in the cultural domain. The two sciences are thus concerned with the same subject area and they have the same ontological foundations. Economic history traces its origins to various roots, in particular to the diverse currents of historical science and the application of economic theories and methods to historical topics. Evolutionary economics can also be regarded as mainstream economics extended by the complexity of endogenous novelty development. Thus, it appears to be a particularly promising approach for research in economic history.3 111

Andreas Resch

Conversely, considerations about specific aspects of evolutionary economics can be derived from the broader tradition of economic history, such as relevant subject matters, fashions of propositions, time dimensions, or the relevance of singularities and generalizable findings. Currently, the focus of evolutionary economics is on neo-Schumpeterian research on firms and industrial dynamics in capitalist market economies. Both historicist and structuralist research strands in economic history encourage to include the social embeddedness of the economy in a broader context of institutional/cultural/political4 development. In this sense, the perceived separation of the study of economic and cultural evolution should be overcome to consider processes of coevolution between these areas. For this purpose, evolutionary grounded institutional economics appear as a particularly suitable approach. This would moreover allow addressing relevant deficits of the prevailing neoinstitutional economics as a historical approach. So far, neoinstitutional economics refers to neoclassics and its ontological basis. With respect to economic history, this leads to two shortcomings. On the one hand, the main question of the neoinstitutional school refers only to whether institutions promote or inhibit development. Yet, the processes of development themselves remain vague; they are merely implicitly assumed to be evolutionary. On the other hand, neoinstitutional research in its narrow interpretation ignores the fact that institutions themselves also emerge in a historical-evolutionary way. For example, Williamson (2000) explicitly states that the subject of his discipline only encompasses the temporal dimension necessary to choose governance structures, but not the long-term periods required for the genesis of institutions. Yet, from the point of view of ‘historically’ understood economic history it seems obvious to engage in sufficiently long-time spans to include evolutionary processes of institution formation in evolutionary economics and thus also to target the coevolution between economy and institutions/culture/politics. In some respects, this would mean a return to positions already claimed by the old American institutionalist school and would be compatible with the neo-Schumpeterian concept of innovation systems. The bestknown representative of American institutionalism, Thorstein Veblen, has himself referred to the German historical school of national economics in this respect (Hodgson 2001, Chapter 10). The innovation system concept was developed with reference to Friedrich List, who in turn developed his theories in the first half of the 19th century under the influence of German historicism. Douglass C. North in his late work has (gradually) adopted the endogenization of institutional development, after having previously profiled himself as the author of a ‘neoclassical’ theory of the state (Wallis 2014). Elinor Ostrom can already be considered longer than North as a scholar in whose work the endogenous development of institutions occupies a central place (Ostrom 1990, 2005). An integrative approach to the economic, political, and cultural evolution of this kind would correspond exactly to what Esben S. Andersen has elaborated from Schumpeter’s work as ‘Schumpeter Mark III’ (Andersen 2011, Chapter XV), i.e. a broad economic-sociological-historical theory of economic and social development. Such an approach also overcomes the thematic limitation of previous neo-Schumpeterian research to developments in capitalist market economies and opens evolutionary economics to important questions of long-term development from traditional societies to different variants of industrial and post-industrial societies. For the discussion of the relevant time dimensions and modes of explanation, very different positions are taken both in evolutionary economics and in (economic) history, depending on the respective question at hand. The mentioned position that in the sense of 112

Evolutionary economics and economic history

the ‘continuity hypothesis’ biological and cultural evolution should be seen in a common domain, which requires very long-time horizons, would be taken with caution by historians. Historical science explicitly separated itself from natural history as early as the 18th century and specialized in human-made developments, but not in human biological developments. Moreover, historians have limited themselves to the study of human history since humanity has produced written sources, that is, for a few thousand years which is an extremely short period compared to the duration of the biological development of the human species. Furthermore, experiences with misguided currents, such as social Darwinism and biologism in the late 19th century, would strengthen this scepticism. From a ‘historical’ perspective, the demarcation strategy that Gul and Pesendorfer (2005) apply for canonical economics vis-à-vis neuroeconomics might seem suggestive for the topic of the repercussions of biological evolution into the cultural sphere. This would mean not including the questions, processes, and methods of the other science in one’s own field, but at most taking into account the results observable and relevant from one’s own perspective. A demarcation of cultural evolution from biological evolution should also prevent dabbling in the latter area. Comparable disputes about the adequate duration of the periods and domains under consideration also occurred, for example, within the framework of the French Annales school. On the one hand, Fernand Braudel (1958, 1995) argued with his concept of the ‘longue durée’ that long-term, structural conditions should be taken as the ultimately most important explanations for historical processes and conditions, against which medium-term cyclical history and short-term event history take place. In theoretical writings, he dismissed event history as a surface phenomenon of little historical interest. Braudel’s primacy of structural downward causality can be countered by the fact that often (temporally) closer causes are more relevant for observed effects (Lorenz 1997, 333f). Marc Bloch, himself one of the founders of the Annales school, also polemicized in this direction against the ‘idol of origins’ in historiography. He warned not to confuse ‘origin’ with ‘cause’ and rather to examine the time in which phenomena occurred in order to explain them (Bloch 2004, 24–29). Moreover, an analysis of Braudel’s opus magnum, The Mediterranean (1995), has revealed that he himself by no means dogmatically paid homage to a primacy of long-term, ‘ultimate’ explanations, but that he was very flexible in incorporating subjective experiences and perspectives as well as coincidence and contingency (Hoffmann 2005). The approach of obtaining ultimate explanations by reconstructing, particularly longterm developmental strands has further to be countered by the fact that both evolutionary economics and economic history do not assume deterministic long causal chains, but rather open processes with contingency, which also gives greater weight to explanatory factors that are closer in time and space. Yet, non-ergodicity, bifurcations, and long-term paths must not be ignored. The multi-layered and thus history-friendly micro-, meso-, and macro-concept of Dopfer and Potts allows to structure different development speeds and thus complex sequences for the differentiated levels. The new occurs through the emergence of new ‘rules’ on the micro and meso levels; thus, overall, a stronger upward than downward causality is assumed, with the micro processes proceeding most rapidly, which is why short- to medium-term processes must be given appropriate attention in this theoretical environment. From all these considerations, it can be deduced that evolutionary economics and economic history always choose a time horizon that is no longer than is plausibly necessary for the question at hand. In this context, cultural and macro processes will tend to define the maximum and micro-trajectories of new ‘rules’ the minimum of meaningful time horizons. 113

Andreas Resch

Finally, different forms of statements about singular events, classes of events, and generally valid statements are discussed. From a common social science perspective, statements about types of events and generally valid statements are most relevant, while, for instance, Witt and Chai (2019) explicitly categorize the concern with singular, historically unique events as irrelevant. This puts them in line with many economists and other social scientists. For example, Popper has noted that only situational logic would be applicable to singular events, which is not of overarching/general interest (Popper 2002). Yet, this view would be countered by encouragement from the historical sciences (especially humanities and poststructuralist approaches) not to ignore the level of single events entirely. For example, individual studies on innovations, case studies on the history of companies or industries, and qualitative studies of diverse individual events definitely serve as a level of reflection for more general concepts as well as potential sources for insights that are of interest in their own right and can be evaluated for generalizability.

Notes 1 Unless otherwise stated, this chapter is based on the following works: Cantner 2016, Dopfer 2011, 2016, Dopfer, Potts 2008, Foster, Metcalfe 2001, Hodgson 2011, Neild 2017, Herrmann-Pillath 2013, Loasby 2019, Metcalfe 2005, Nelson et al. 2018, Witt 2001, 2008, Witt, Chai, 2019. Specifically addressing ontological aspects are, for example, Dopfer 2005, Herrmann-Pillath 2001, Hodgson, 2011, Robert et al. 2017, and Vromen 2019. 2 For the following discussion, individual references will again be largely dispensed with. The literature on the topics addressed has become almost unmanageable, which is why only a few handbooklike introductions are mentioned here: Koselleck 1975, Lorenz 1997, Iggers et al. 2016, Woolf, 2011, and Jordan 2013. 3 See also Dopfer 2001, Mokyr 2005, 2019. 4 Recently, an entire special issue of the Journal of Evolutionary Economics was dedicated to the topic of ‘Evolutionary Economics and Policy’. For the introduction to this volume see Brenner, Broekel 2019. For content and methods of evolutionary political economy see also Hanappi, Scholz-Wäckerle 2017.

References Andersen E S (2006), Appraising Schumpeter’s ‘Essence’ after 100 Years: From Walrasian Economics to Evolutionary Economics, Druid Working Paper 06–35, Danish Research Unit for Industrial Dynamics: Aalborg. Andersen E S (2011), Joseph A. Schumpeter. A Theory of Social and Economic Evolution, Palgrave Macmillan: Basingstoke et al. Bloch M (2004), The Historian’s Craft, Manchester University Press: Manchester and New York, NY. Bowles S, Gintis H (2011), A Cooperative Species. Human Reciprocity and Its Evolution, Princeton University Press: Princeton, NJ, et al. Boyd R, Richerson J. (1985), Culture and the Evolutionary Process, The University of Chicago Press: Chicago, IL. Braudel F (1958), Histoire et Sciences sociales. La longue durée. Annales. Économies, Sociétés, Civilisations 13(4), 725–753. Braudel F (1995), The Mediterranean and the Mediterranean World in the Age of Philipp II, 3 Volumes, University of California Press: Berkeley, CA, and Los Angeles, CA. Brenner Th, Broekel T (2019), Evolutionary economics and policy: Introduction to the special issue. Journal of Evolutionary Economics 29, 1373–1378. Cantner U (2016), Foundations of economic change – an extended Schumpeterian approach. Journal of Evolutionary Economics 26, 701–736.

114

Evolutionary economics and economic history Cordes Ch (2019), The promises of a naturalistic approach: How cultural evolution theory can inform (evolutionary) economics. Journal of Evolutionary Economics 29, 1241–1262. Coyle D (2018), In defence of the economists. In Prospect https://www.prospectmagazine.co.uk/ economics-and-finance/dianecoyle (Dec. 30, 2020). Dopfer K (2001), History-friendly theories in economics: Reconciling universality and context in evolutionary analysis. In J Foster, J S Metcalfe (eds.), Frontiers of Evolutionary Economics. Competition, Self-Organization and Innovation, Edward Elgar: Cheltenham and Northampton, MA, 160–194. Dopfer K (2005), Evolutionary economics: A theoretical framework. In K Dopfer (ed.), The Evolutionary Foundations of Economics, Cambridge University Press: Cambridge et al., 3–59. Dopfer K (2011), Economics in a cultural key: Complexity and evolution revisited. In J B Davis, D Wade Hands (eds.), The Elgar Companion to Recent Economic Methodology, Edward Elgar: Cheltenham and Northampton, MA, 319–340. Dopfer K (2016), Evolutionary economics. In G Faccarello, H D Kurz (eds.), Handbook on the History of Economic Analysis, Vol III, Developments in Major Fields of Economics, Edward Elgar: Cheltenham and Northampton, MA, 175–193. Dopfer K, J Potts (2004), Micro-meso-macro: A new framework for evolutionary economic analysis. In J S Metcalfe, J Foster (eds.), Evolution and Economic Complexity, Edward Elgar: Cheltenham. Dopfer K., J. Potts (2008), The General Theory of Economic Evolution, Routledge: London and New York, NY. Foster J, Metcalfe J S (2001), Modern evolutionary economic perspectives: An overview. In J Foster, J S Metcalfe (eds.), Frontiers of Evolutionary Economics. Competition, Self-Organization and Innovation Policy, Edward Elgar: Cheltenham and Northampton, MA, 1–16. Gul F, W Pesendorfer (2005), The Case for Mindless Economics, Princeton University, https://www. princeton.edu/~pesendor/mindless.pdf (Dec. 30, 2020). Hanappi H, Scholz-Wäckerle M (2017), Evolutionary political economy: Content and methods. Forum for Social Economics, 50, 1–17. Henrich J (2016), The Secret of our Success: How Culture is Driving Human Evolution, Domesticating our Species, and Making us Smarter, Princeton University Press: Princeton, Oxford. Hawkins M (1997), Social darwinism in European and American thought 1860–1945. Nature as Model and Nature as Threat, Cambridge University Press: Cambridge, New York, NY, and Melbourne. Herrmann-Pillath C (2001), On the ontological foundations of evolutionary economics. In K Dopfer (ed.), Evolutionary Economics: Program and Scope, Kluwer Academic Publishers: Boston, Dordrecht, and London, 89–139. Herrmann-Pillath C (2013), Foundations of economic evolution. A Treatise on the Natural Philosophy of Economics, Edward Elgar: Cheltenham. Hodgson G M (2001), How economics forgot history. The Problem of Historical Specificity in Social Science, Routledge: London and New York, NY. Hodgson G M (2011), A philosophical perspective on contemporary evolutionary economics. In J B Davis, D Wade Hands (eds.), The Elgar Companion to Recent Economic Methodology, Edward Elgar: Cheltenham and Northampton, MA, 299–318. Hoffmann A (2005), Zufall und Kontingenz in der Geschichtstheorie, Vittorio Klostermann: Frankfurt/Main. Iggers G G, Q E Wang, S Mukherjee (2016), A Global History of Modern Historiography, 2nd Edition, Taylor & Francis: United Kingdom. Jordan S (2013), Theorien und Methoden der Geschichtswissenschaft. 2. Auflage, Schöningh UTB: Paderborn. Koselleck R (1975), Die Herausbildung des modernen Geschichtsbegriffs. In Geschichtliche Grundbegriffe, Band 2 E–G, Ernst Klett Verlag: Stuttgart, 647–706. Loasby B J (2019), Missed connections and opportunities forgone. In U Witt, A Chai (eds.), Understanding Economic Change, Cambridge University Press: Cambridge et al., 43–80. Lorenz Ch (1997), Konstruktion der Vergangenheit. Eine Einführung in die Geschichtstheorie, Böhlau: Köln, Weimar, and Wien. Mesoudi A (2011), Cultural Evolution, University of Chicago Press: Chicago. Metcalfe J S (2005), Evolutionary concepts in relation to evolutionary economics. In K Dopfer (ed.), The Evolutionary Foundations of Economics, Cambridge University Press: Cambridge, Melbourne, and Cape Town, 391–430.

115

Andreas Resch Mokyr J (2002), The Gifts of Athena: Historical Origins of the Knowledge Economy, Princeton University Press: Princeton, NJ. Mokyr J (2005), Is there a theory of economic history? In K Dopfer (ed.), The Evolutionary Foundations of Economics, Cambridge University Press: Cambridge et al, 195–218. Mokyr J (2018), A Culture of Growth. The Origins of the Modern Economy, Princeton University Press: Princeton, NJ. Mokyr J (2019), Science, technology, and knowledge. What economic historians can learn from an evolutionary approach. In U Witt, A Chai (eds.), Understanding Economic Change, Cambridge University Press: Cambridge, 81–119. Neild R (2017), The future of economics: The case for an evolutionary approach. The Economic Labour Relations Review, 28(1), 164–172. Nelson R R et al. (eds.) (2018), Modern Evolutionary Economics, Cambridge University Press: Cambridge et al. Nelson R, Winter S (1982), An Evolutionary Theory of Economic Change, Harvard University Press: Cambridge, MA. Nunn, N (2020), History as Evolution, Working Paper 27706, National Bureau of Economic Research: Cambridge, MA. Ostrom E (1990), Governing the commons. The Evolution of Institutions for Collective Action, Cambridge University Press: Cambridge. Ostrom E (2005), Understanding Institutional Diversity, Princeton University Press: Princeton, NJ. Peneder M (2017), Competitiveness and industrial policy: From rationalities of failure towards the ability to evolve. Cambridge Journal of Economics 41, 829–858. Popper K (2002), The Poverty of Historicism, Routledge Classics: London and New York. Richerson P J, M H Christiansen (eds.) (2013), Cultural Evolution: Society, Technology, Language, and Religion, The MIT Press: Cambridge, MA, and London. Robert V, G Yoguel, O Lerena (2017), The ontology of complexity and the neo-Schumpeterian evolutionary theory of economic change. Journal of Evolutionary Economics 27, 761–793. Vromen J (2019), Generalized darwinism in evolutionary economics. The devil is in the detail. In U Witt, A Chai A (eds.), Understanding Economic Change, Cambridge University Press: Cambridge et al., 120–154. Wallis J J (2014), Persistence and change in institutions. The evolution of Douglass C. North. In S Galliani, I Sened (eds.), Institutions, Property Rights, and Economic Growth. The Legacy of Douglass North, Cambridge University Press: New York, NY, 30–49. Williamson O E (2000), The new institutional economics: Taking stock, looking ahead. Journal of Economic Literature 38(3), 596–600. Witt U (2001), Evolutionary economics: An interpretative survey. In K Dopfer (ed.) Evolutionary Economics: Program and Scope, Kluwer Academic Publishers: Boston, Dordrecht, and London, 45–88. Witt U (2008), What is specific about evolutionary economics? Journal of Evolutionary Economics, 18, 547–575. Witt U (2013), The Future of Evolutionary Economics: Why Modalities Matter, Papers on Economics & Evolution, # 1309, Max Planck Institute of Economics, Evolutionary Economics Group: Jena. Witt U, A Chai (2019), Evolutionary economics. Taking stock of its progress and emerging challenges. In U Witt, A Chai (eds.), Understanding Economic Change. Advances in Evolutionary Economics, Cambridge University Press: Cambridge et al., 3–40. Woolf D (2011), A Global History of History, Cambridge University Press: Cambridge.

116

9 WHY AN EVOLUTIONARY ECONOMIC GEOGRAPHY? The spatial economy as a complex evolving system Ron L. Martin and Peter J. Sunley

9.1

Introduction: The geographical foundations of the economy

It is probably not too unfair to suggest that economics has never really taken geography seriously. This is true both for mainstream orthodox (neoclassical) economics and heterodox economics. In both schools, among the vast majority of economists, the economy has been traditionally viewed as consisting of two levels, the macro- (national) economy of aggregate trends and conditions, and the micro-economy of individuals and firms, with no stopping place in between. While Hodgson (2001) has complained that economics has tended to forget history, it can be equally argued that it has tended to forget geography, or place and location. In mainstream economics, for example, the focus has been on developing theoretical models of macro- or micro-economic processes that are assumed to hold across all times and all places. The quest for universality has taken precedence over considerations of historical or geographical context and specificity, while the overriding assumption that economies naturally settle into equilibrium states severely limits any real interest in history and dynamics (other than movement from one equilibrium state to another). Within the heterodox camp, evolutionary economics does at least seek to remedy the neglect of history by assigning central importance to how firms, industries, and institutions actually evolve through time, and in contrast to mainstream economics works with the assumption that economic systems are continually in disequilibrium. However, evolutionary economics, no less than its mainstream counterpart, has also given little attention to geography and its role in shaping how firms, industries, technologies, and institutions evolve.1 To be sure, there have been notable exceptions to this lack of a geographical lens in economics. After all, it was Alfred Marshall (1919), one of the founding fathers of mainstream economics, who coined the notion of ‘industrial districts’ in recognition that industrial development is a quintessentially local process2; while the heterodox economist Gunnar Myrdal (1957) later elaborated his idea of ‘circular cumulative causation’ in relation to the tendency for economic development to be regionally uneven and disequilibrating. And although the two subfields of regional economics and urban economics, which do address the geographical dimensions of economic activity and development, have existed for well over half a century, and have become sophisticated branches of theoretical

DOI: 10.4324/9780429398971-11

117

Ron L. Martin and Peter J. Sunley

and empirical work over the past two or three decades, they have remained marginal to the economics discipline as a whole. Even the recent rise of the so-called ‘new economic geography’ (NEG), pioneered by Paul Krugman, which builds on Marshall’s notion of localisation economies and the idea of increasing returns based on cumulative causation, together with elements of new trade theory, to construct formal models of equilibrium economic landscapes, has yet to inject a ‘spatial imaginary’ into the economics discipline more generally. For economic geographers, however, ‘the economy’ is necessarily a spatially situated and organised system. For them, the specificity of place is not simply some ‘complicating factor’ to be left out of models of the economy, but is intrinsic to how economic processes operate: geography matters. To adapt one of Marshall’s own phrases, it is in individual localities, cities, and regions that the ‘everyday business of economic life’ is carried out, where firms are located, where they invest and innovate, where goods and services are produced and consumed, where trade is conducted, where workers are employed and wages are paid, and where public services are delivered and accessed. The economy is not some ‘average’ construct or aggregate entity, but a composite and multi-scalar landscape, a spatial mosaic which at any point in time is the product of past investments and disinvestments in individual places by firms and the state, that in turn presents firms and the state with a fresh surface of local opportunities (and constraints) for future investment and disinvestment. The spatial economy, in other words, is a complex evolving system, in which multiple locally differentiated contexts and multiple temporal rhythms determine the pace, nature, and outcomes of evolutionary economic processes. Capitalism cannot stand still. Its central imperative – the search for profitable wealth creation – drives a perpetual process of economic flux. Joseph Schumpeter famously called this a process of ‘creative destruction’.3 The two sides of this process are neither geographically even in origin nor neutral in impact. Over the years, geographers have built up an extensive body of analysis of what they often refer to as ‘uneven geographical development’, or the inherent tendency under capitalism for economic growth and development to favour some areas and regions rather than others. Typically, they have conceptualised economic landscapes as the outcomes, at least in part, of the tensions between opposing dualistic processes – of centripetal forces of spatial agglomeration versus centrifugal processes of spatial dispersal; of forces making for spatial homogenisation versus those making for geographical differentiation; and of forces of regional convergence versus those of regional divergence. Over the past two decades, a new paradigm has emerged and developed within economic geography concerned with conceptualising and explaining the spatial development and organisation of economies in explicitly evolutionary terms (see, for example, Boschma and Frenken, 2011; Boschma and Martin, 2010; Kogler, 2017; Martin and Sunley, 2015).4 This ‘evolutionary turn’ is different and distinctive in the primacy it seeks to give to the forces that determine the nature, pace, and direction of change in economic landscapes over real historical time. An evolutionary perspective adds another key dualistic tension, between processes and factors making for continuity in the spatial configurations of economic activity, and those making for change and transformation in those configurations. Just as the rise of evolutionary economics seeks to remedy the traditional neglect of history in modern economics, and to highlight how economies are transformed over time, so evolutionary economic geography goes further by demonstrating that many of the processes that drive economic evolution are inherently geographical in origin and operation. To that end, evolutionary economic geographers have naturally drawn on the ideas from evolutionary economics, including the key precepts from 118

Why an evolutionary economic geography?

Generalised Darwinism (variety, selection, and retention), but also from path dependence theory, institutional theory, and ideas from complexity theory, on adaptability, emergence, and resilience. In their original form, most of those ideas ignore or neglect geography. Thus, a key task in evolutionary economic geography has been to argue how evolutionary economic processes are inherently spatial in nature and how geography both shapes those processes and is produced by them. In so doing, not only is the use and application of ideas from evolutionary economics helping geographers to understand geographically uneven development, but at the same time evolutionary economic geography can help in elaborating and refining those very ideas themselves.

9.2

The scope and contribution of evolutionary economic geography

There are two basic ways of summarising the scope of evolutionary economic geography: the different theoretical and conceptual tools used, and the range of applications to empirical spatial economic phenomena (see Figure 9.1). Evolutionary economic geographers have applied a range of evolutionary concepts and approaches to a set of well-established research questions in economic geography, including the causes and processes of clustering and agglomeration, the growth and decline of cities and regions, and the spatial distribution of technological innovations and new industries (Boschma and Frenken, 2018). As Figure 9.1 also shows, a given theoretical or conceptual approach has frequently been used to investigate more than one empirical issue. Thus, for example, the notion of path dependence (and its dual, path creation) has been used to develop explanatory accounts of the evolution of particular industries across geographic space, the evolution of local business clusters, the industrial histories of particular case-study regions (and cities), and change and continuity of local institutional forms and structures. Thus far, evolutionary economic geography has been a loose and pluralistic framework which has combined both old and new ideas (Boschma and Martin, 2010; Coe, 2011). Nevertheless, it has typically emphasised three key themes or principles: history, heterogeneity, and holism. These principles are implicit or explicit in most of the areas shown in Figure 9.1 and together they point to some of the ways in which processes of economic change and continuity are inherently spatial. If we are to properly understand the significance of history, heterogeneity, and holism in economic evolution, then we need to examine how they are materialised and effected through place and geography.

9.2.1 Three key guiding principles The emphasis on history stems from the core belief that an understanding of spatial economic landscapes ultimately rests on explaining how they have unfolded over time and how they have come to be (Hall, 1962; Boschma and Martin, 2010; Martin and Sunley, 2022). Thus the variety of conceptual approaches found in evolutionary economic geography, ranging from Generalised Darwinism to complex systems theories, share a common concern with historical process (Boschma and Frenken, 2018). As Henning (2019, p. 1) writes ‘Like few other approaches, evolutionary economic geography recognises the importance of both time and history to a scientific understanding of regional development’. Its key distinguishing feature is that it promises to provide a conceptual approach that shows how history matters for urban and regional development, where history is understood broadly as an empirical and/or theoretical concern with and/or use of the past. This includes both using 119

Ron L. Martin and Peter J. Sunley Theoretical/Conceptual Framework

Topic of Empirical Application

Generalised Darwinism

Evolution of Industries across Space

Path Dependence and Path Creation Theory

Cluster Life Cycles and Evolution

Complexity and Emergence

Regional industrial Change and Adaptation

Related Variety and Diversification

Spatial Evolution of Knowledge and Networks

Network Theory and Relational Theory

Changing Geographies of Innovation

Adaptive Cycle Theory and Panarchy

Evolution of Local Institutional Forms

Schumpeterian Innovation Theory Darwinism

Resilience of Regional and City Economies to Shocks

Resilience Theory

Regional Energy Transitions

Figure 9.1 The scope of evolutionary economic geography: Concepts and applications.

historical data to modify, test and develop geographical theories, and developing theories that see the past as a causal influence on the future course of change. While the use of longitudinal data and historical methods has been slow to develop (Martin and Sunley, 2015; Henning, 2019), as we will see, another goal has been to understand how inherited spatial patterns and legacies shape new investments and the behaviour and opportunities of economic agents and firms. This is not to say, however, that evolutionary economic geography has fully exploited the explanatory scope that different modes of historical causal investigation offer, and there is considerable scope for making history matter more (Martin and Sunley, 2022). 120

Why an evolutionary economic geography?

Heterogeneity, or the presence and production of diversity and variety, is of course fundamental to evolutionary theories. The continual creation of economic variety, the competitive selection and winnowing of new economic elements (firms, innovations, industries, and products), and the retention of successful entities and practices are basic evolutionary economic processes. Evolutionary economic geography has reflected this by examining how variety creation, through innovation, branching, and other diversification processes, varies over space, and by showing how economic heterogeneity is fundamental to the evolution of economic-geographical entities such as business clusters, cities, and regions. Much of this work has sought to understand the constituent parts and populations of geographical entities and to examine how changes in these components are related to the evolution and development of the entity as a whole. Several different types and dimensions of heterogeneity are recurrent. Some have looked at variations in actors such as entrepreneurs. Most prominently, many studies have stressed the importance of differences between firms and their capabilities and resources, and have examined the ways in which these are central to geographical change. These have often been traced to the variety of intangible assets and tacit knowledge available within individual regions (see Boschma, 2004), although more recently transfers of extra-regional knowledge, and the absorption of exogenous resources, have been shown to be key mechanisms of variety and change (Trippl et al., 2018). Others have focused on technological paths and trajectories, defined as a fusion of technologies with established ways of problem-solving and dominant practices. Such trajectories are followed by a group or cluster of firms, or even by an entire industry. Yet another set of studies has highlighted industries as the key constituents and focused extensively on industrial heterogeneity. The third key principle is that of holism which has been implicit in much work in evolutionary economic geography, even though it has been less explicitly invoked. In essence, it refers to the idea that the parts of any entity are intimately connected and that wholes are more than the sum of their parts. This is, of course, a key characteristic of complex adaptive systems, which undergo phases of structural change and re-organisation that are irreversible (Martin and Sunley, 2007, 2012; Rigby, 2018). Much of this is applicable to local and regional economies which emerge from interactions between their parts, but which also then influence how these components (such as firms and workers) change and develop. Local and regional economies thus act as key contexts in which relationships between economic agents and institutions co-evolve and shape the nature of economic activity (Essletzbichler and Rigby, 2007; Storper, 2009). Accordingly, there has been growing interest in the way in which individual agency is enabled or limited by systemic conditions, such as regional innovation and entrepreneurial ‘ecosystems’ (for example, see Isaksen and Trippl, 2017; Stam, 2015). This perspective highlights the importance of emergence, whereby certain outcomes, patterns, and paths emerge from, but cannot simply be reduced to, the behaviours of the micro-units of which the system in question is composed (Martin and Sunley, 2012).5 Emergent macro-outcomes, patterns, and paths have downward causative influences on those micro-units. Moreover, such upward and downward processes operate across different scales and with different temporalities (from slow, incremental and cumulative processes to periodic cyclical rhythms to episodic ruptures and disruptions (Martin and Sunley, 2022)),6 so that while urban and regional economies act as important holistic contexts for economic processes, they are not bounded and closed entities but are entwined and integrated with broader national and global forces and relationships (Essletzbichler, 2009). Just as temporal processes shape geographical outcomes so those outcomes may influence 121

Ron L. Martin and Peter J. Sunley

temporal processes. The economic landscape consists of a layered ontology in which multiscale relationships are complex and interactive. This spatial heterogeneity in the pace and nature of economic change and transformation will be specific to and contingent on the historical context in which it unfolds. But, furthermore, the very outcomes in individual regions and places can feed back to influence the temporalities of capitalist development and evolution. As Doreen Massey emphasised, just as the temporalities of the economy are needed to understand its spatialities, so the latter can in turn influence those temporalities: Spatial form as ‘outcome’ … has emergent powers which can have effects on subsequent events. Spatial form can alter the course of the very histories that have produced it … One way of thinking about all of this is to say that the spatial is integral to the production of history … just as the temporal is to geography. (Massey, 1992, p. 84) This recursive relationship between the temporal and the spatial seems to us to be a crucial feature requiring the attention of evolutionary economic geographers.

9.2.2 Putting the principles in place The effects and outcomes of history, heterogeneity, and holism are profoundly interdependent and recursive. Historical legacies and past assets are often causes of economic heterogeneity, and it is the degree and nature of this heterogeneity and its interactions with other place-specific contextual conditions, which then drive further economic development. All three of these principles are inherently geographical as the key mechanisms through which they operate are shaped by local context, as well as being stretched across different spaces and scales. For example, the importance of firm heterogeneity has been a recurrent theme in evolutionary studies of clusters. A substantial number of cluster and entrepreneurial studies have drawn on organisational ecology and reported that spin-off firms tend to remain geographically close to their parent firms and that this localised dynamic can lead to the formation of business clusters (see Buenstorf and Klepper 2009; Stam, 2009). Moreover, spin-offs typically inherit routines and experience from the parent firms and therefore enjoy a higher survival rate than firms without these inheritances (Boschma and Frenken, 2011). In place of generalised Marshallian externalities, such heredity effects mean that while some firms may benefit from being located in a cluster, others may be weakened by competition and congestion effects (Frenken et al., 2015; Rigby and Brown, 2015). Heterogeneity has also been seen as a key determinant of the life cycle of clusters where it has been argued that the loss of knowledge heterogeneity through myopia and cognitive ‘groupthink’ may lead to a decline of particular clusters. This suggests that the renewal of knowledge heterogeneity is key to the survival of industry clusters (Maskell and Malmberg, 2007) so that the attraction of new agents and entrepreneurs and absorption of external knowledge can refuel and revitalise clusters (Martin and Sunley, 2011). Evolutionary economic geography has given substantial attention to the connections between firm heterogeneity and firms involvement in different types of knowledge networks. The knowledge held and used by firms often stems from their integration into different types of networks and relationships with other firms and agents, and these networks show marked 122

Why an evolutionary economic geography?

geographical patterns. They are shaped by distance and transmission costs, as well as by their institutional settings, and they themselves evolve over time (Gluckler, 2007; Boschma and Frenken, 2011). This work has sought to understand why some firms are more centrally located in networks, and why they are better able to absorb knowledge flows. It has made some progress in understanding the relative importance of geographical proximity in different forms of knowledge sharing. The role of different types of formal and informal networks in supporting, or restricting, knowledge sharing between economic agents has been found to be a key reason why some cities and regions prove more dynamic and productive than others (Balland et al., 2016; Storper et al., 2015). Recent analyses of urban differences in productivity have also found that firm heterogeneity within industries is a key cause of variations in productivity across cities, and have thereby raised a further set of questions about the causes of firm capabilities and their relationships to place (Martin et al., 2018). There is evidently much scope for further work into the dimensions and causes of firm and network heterogeneity across space (Rigby, 2018). A related set of studies has argued that the exhaustion and decline of knowledge heterogeneity can undermine the prosperity and growth of local and regional economies. It highlights the risks of path dependence and ‘lock-in’ to a single technological path. The classic notion of path dependence provided an ‘off-the-shelf’, ready-made explicit model of historical process as following a predictable sequence. In this sequence, early decisions and accidental events have long-term, probable consequences that narrow options and may eventually lead to ‘lock-in’ and an inability to escape from a familiar industrialtechnological path. In most cases, path dependence has been applied to the rise and fall of clusters and industrial districts over the long run and certainly captures some important aspects of increasing returns and cumulative change in regional economic development. In some cases, path dependence operates through place-dependence (Martin and Sunley, 2006). While this certainly broadens our gaze away from micro-scale processes and shows that economic trajectories are fundamentally historical, it only provides a rather limiting view of historical sequence (Henning et al., 2013). Its limited understanding of cumulative causation means that it tends to neglect the ways in which past conditions are just as often enabling as they are constraining. The theory tends to ignore other sequences of continual incremental diversification and renewal (Martin, 2010). More refined analyses of different paths, including path extension, upgrading, branching, diversification, and new path creation, have illuminated a wider range of possible trajectories of change (Isaksen and Trippl, 2017). In some cases, path dependence has been integrated within concepts of cluster and industry life cycles that attempt to distinguish key regularities in the past that act as a guide or template for analysing change (Menzel and Fornahl, 2010). However, there are many unresolved questions about the strength and generalisability of this life cycle model (Martin and Sunley, 2011; Trippl et al., 2015) and, indeed, whether regular life cycles witnessed in industrial history continue to apply to post-industrial knowledge-based forms of capitalism. In recent years, studies in evolutionary economic geography have switched from an interest in classic path dependence to focus much more on path creation and the creation of new technological pathways (MacKinnon et al., 2019; Grillitsch and Sotarauta, 2020; Isaksen and Trippl, 2017). Some of this work focuses on the renewal of heterogeneity and diversity through recombination and suggests that a particular form of heterogeneity or related variety is especially beneficial (Frenken et al., 2007). This is based on a recombinant model of innovation which argues that innovations occur mainly through agents recombining and mixing ideas from cognitively related fields. A certain degree of cognitive 123

Ron L. Martin and Peter J. Sunley

proximity between industries is necessary for them to learn from each other. In some cases, regional economies may benefit from a radical departure into previously unrelated technological fields and industries, but such departures are seen as rare and hard-to-replicate events (Boschma, 2017). The key process envisaged by this relatedness literature is one of knowledge and capability sharing. This key engine is seen as driving a wide variety of forms of regional economic growth and has been argued to lead to a wide range of benefits including higher employment growth, greater resilience, higher productivity, more innovation, and new technological path creation (Neffke et al., 2011). The past shapes the course of geographical economic change through spatially variegated legacy effects, which provide an enabling menu of assets and resources available in particular places. Recent work has typically used examples of industrial diversification in the past, such as in the Basque Country or Lyon, as evidence for this process of related diversification (for example, Boschma et al., 2017). There has been a growing number of empirical studies that have attempted to assess whether ‘related variety’ does in fact lead to faster rates of growth and innovation. Empirical verification and support for these claims have struggled to keep up with the fast pace of enthusiasm for the concept, however. Because ‘relatedness’ is difficult to measure directly and unambiguously, empirical measurement of relatedness has been forced to rely on a number of proxies, which have produced varied and mixed results (see Aarstad et al., 2016; Fritsch and Kublina, 2018; Castaldi et al., 2015). Different forms of relatedness such as product complexity, patent co-citations, labour flows, SIC categories, and entropy measures have all been used and may not be measuring the same types of knowledge relatedness (for example, see Cicerone et al., 2019; Boschma and Iammarino, 2009). The precise meaning of ‘variety’ has often not been carefully defined so there is much scope for further critical analysis of how firm capabilities and knowledge recombination may produce industrial diversification and regional economic renewal. In many cases, ‘related variety’ is presented as a universal theory of industrial growth and diversification, and in stressing the benefits of this form of diversity some studies have lost sight of the importance of holism and history. Studies have tended to assume that the key processes and mechanisms inevitably operate, irrespective of institutional settings, historical periods, and indeed regional context. There is little doubt that industrial diversification is an important industrial dynamic, but there has been little explanation of why and how it varies across time and space. The timescale and operation of these diversification processes require further research. More holistic approaches have shown that innovation dynamics are necessary but not sufficient for regional path creation. Instead, global firms and production systems, regional innovation systems, and policy architectures need to be coupled and aligned effectively in order to create the conditions for the development of a new industry (for example see Chlebna and Simmie, 2018; Coenen et al., 2015; Dawley, 2014; MacKinnon et al., 2019). Individual entrepreneurship, institutional entrepreneurship, and place leadership are often all implicated (Grillitsch and Sotarauta, 2020). History is once again important, however, as regional institutional capabilities that have supported earlier industries may show a certain plasticity and may gradually be moved to the support of new industries and sectors (Strambach, 2010). The rapid expansion of research into urban and regional economic resilience has also underlined the significance of heterogeneity, history, and holism. The analysis of how cities and regions respond to the shocks and impacts of recessions has sought to explain why some places recover far more rapidly and successfully than others. It has given substantial 124

Why an evolutionary economic geography?

attention to the proposition that more diverse economies are more resilient than more specialised and homogenous economies. In some instances, this has proved to be the case (for example, see Evans and Karecha, 2014). However, recent work has found that industrial structure is not always the most important determinant of resilience, and there are also significant variations between firms in the same sectors and industries across cities and regions (Martin et al., 2016; Martin and Gardiner, 2019, 2020). Firm heterogeneity appears to be a complex and significant factor that is strongly linked to the context of different places. Explanations of resilience have therefore sought to examine a wide range of structural, labour market, and institutional factors and how these interact in the context of particular economies (Sensier et al., 2016; Wink et al., 2016). These studies have argued for an adaptive resilience process which goes beyond seeing resilience simply as recovery to a pre-given trend and takes history more seriously by placing responses to recessions in the context of longer-term structural changes and hysteretic effects. The key focus of analysis should be the recursive interactions between recessions and these structural changes (Simmie and Martin, 2010; Martin and Sunley, 2015a). Resilience appears to change significantly over time as shocks shape urban and regional development paths and as the causes and sources of different shocks change (Martin and Gardiner, 2019). An evolutionary approach to the study of regional and urban economic resilience is called for (see Martin, 2018; Boschma, 2015).

9.3 The evolution of evolutionary economic geography: Future challenges and directions A considerable and impressive body of evolutionary economic geography literature, both conceptual and empirical, has thus been built up over a relatively short time, sufficient to justify taking stock of the direction in which it might – or should – develop over the next couple of decades. A number of challenges that will need to be confronted as the discipline moves forward can be identified, but three will most certainly require attention from evolutionary economic geographers: a Is the current empirical remit of evolutionary economic geography too narrow? Does its field of focus need to be widened? For example, what does evolutionary economic geography have to say on the big processes, large structures, and recurrent crises that appear to be driving 21st-century capitalism? b Does the current lack of a unifying theoretical framework matter? Is it possible, indeed desirable, to construct such a framework? What would that framework look like? Should evolutionary economic geography seek to integrate its ideas with those of other branches of economic geography, such as geographical political economy, with its focus on the unstable dynamics of geographically uneven capitalist development? c Should greater priority be accorded to addressing key spatial economic policy questions and challenges?

9.3.1

Widening the field of focus

With regard to (a), the specific zone of enquiry that distinguishes evolutionary economic geography, although at a broad level it can be argued that the field is concerned with how economic landscapes evolve and change, in practice much of the effort has been directed to two main issues – to the changing geographies of industrial structures and dynamics, and 125

Ron L. Martin and Peter J. Sunley

the geographies of technological change and innovation. In fact, the focus on industrial dynamics has increasingly dominated the field. To be sure, there has been work on the role of institutions in shaping the evolution of industries and technology across geographical space, and some on how institutions themselves have evolved and developed geographically. But the overwhelming thrust of research has been on industrial change. Indeed, evolutionary economic geography has been preoccupied with the micro-level of the firm, although with a welcome, growing, and secondary focus on the meso-level industry structures of regional and urban economies and their institutional configurations and collective agency. As yet, evolutionary economic geography has not really grappled with the spatialities and temporalities of the big processes and large structures of capitalist development in its broader sense, including its crisis tendencies. Evolutionary economic geography has emerged precisely at a time when economic systems have themselves been subject to recurrent phases of disruption and turmoil of historic proportions. The Global Financial Crisis of 2008–2009, though having its proximate causes in the collapse of the subprime mortgage market in the United States and elsewhere, was more fundamentally the result of the highly unbalanced model of economic globalisation that has developed since the late1970s. That imbalance has been quintessentially geographical, a mode of global economic growth led by and benefitting certain countries, regions, cities, and social groups, but leaving other countries, regions, cities, and social groups behind. How to resolve this great imbalance is one of the big problems of our times. The more so, since the global COVID-19 pandemic that swept through much of the world in 2020–2021 has not only had dramatically uneven social and economic impacts geographically, but is almost certain to lead to significant shifts in the structure and organisation of economic activity, in the labour market and nature of work, in the role and functioning of cities, in innovation, and in public policy, all with uneven geographical ramifications. And added to this, there are other challenges, all of which threaten also to be highly geographically uneven in their mechanisms and effects – not least the disruptive effects of new technologies (especially artificial intelligence and robotics), the shift to a green economy, a net-zero carbon economy, instabilities in world trade, the crisis of economic governance (both central government institutions and international organisations), the problem of environmental and resource degradation, and of course climate change. All of these processes and disruptions threaten to intensify the social and spatial inequalities that have developed over recent decades. Evolutionary economic geography has only just begun to apply its methods and conceptual ideas to such big issues, with studies beginning to emerge on the regional energy transition, and the geographies of the so-called ‘Industry 4.0’ revolution (based on robotics, artificial intelligence, and related digital technologies). Of course, one might legitimately respond by arguing that evolutionary economic geography cannot be expected to provide answers to all of these and other big questions and challenges and that its domain of interest is necessarily more narrowly focused. Some have argued similarly in defence of evolutionary economics, that it is not intended to cover the broad sweep of economic development processes and dynamics. But in each case, this merely raises the question of how a more narrowly defined objective, and object of study, relates to and interacts with, other approaches. There has been a debate within economic geography over whether the aim should be a separate body of theory and empirics, labelled ‘evolutionary economic geography’, or whether, rather, the task should be to embed evolutionary ideas, concepts, and processes explicitly within economic geography, within or in 126

Why an evolutionary economic geography?

combination with other explanatory frameworks that characterise the field. A third possibility, of course, is the idea of embedding and integrating ideas and concepts from other approaches used in economic geography into evolutionary economic geography. For example, in the vast literature that has emerged in recent years on cities and their role in economic growth and development, much emphasis has been directed to the role and impact of agglomeration, human capital, specialisation, and economic governance. Now evolutionary economic geography has had, and is having, some interesting things to say on economic specialisation and diversity, but it has had much less to say on the evolutionary dynamics of agglomeration, human capital, or changing governance structures. There may well be considerable scope for further developing evolutionary economic geography on these fronts.

9.3.2 Moving beyond ‘patchwork’ evolutionary economic geography: In search of a unifying theoretical framework? The complaint by some evolutionary economists that their field has failed to produce a coherent and generally agreed unifying framework or narrative carries over to evolutionary economic geography. As summarised in the previous section, evolutionary economic geography now encompasses a range of different key concepts and ideas – path dependence, related variety, relational thinking, networks, institutional change, resilience, adaptability, and several others. But it is difficult to discern the formation of an overarching framework or set of organising principles that link these various ideas and concepts into a coherent theoretical core. Of course, much depends on the view of what such a core should be, what its key object of study should be, or even if the construction of such a core is in fact desirable. At stake here is the issue of theoretical pluralism, and how to manage it. In response to the lack of a unifying framework within evolutionary economics some of its leading proponents – including Ulrich Witt, Geoff Hodgson, Sidney Winter, and JanWillem Stoelhorst – have argued that what is needed is a return to and further generalisation of the core ‘naturalistic’ principles of Generalised Darwinism (variety, selection, and retention or heredity) as the basis of an evolutionary ontology for economics (Stoelhorst, 2014). But Levit et al. (2011) argue that Generalised Darwinism is too ‘top-down’, trying in vain to proceed from an abstract hull to auxiliary hypotheses about economic processes. Instead, they argue, a better strategy might be a ‘bottom-up’ approach that starts with concrete details: The recommendation for evolutionary economics would be to focus on analysing the huge variety of specific evolutionary processes in the economy at the concrete level, and only when explanatory progress has been made at that concrete level to engage in a (bottom-up) discourse of how the complex set of hypotheses can be organised into a more coherent causal and functional structure. (Levit et al., op cit, p. 559) This arguably resonates better with the ‘bottom-up’ or case-study approach followed by many economic geographers, with their concern with the importance of local context and specificity in shaping the nature and operation of economic processes. In contrast, in their search for a unifying ontological framework in evolutionary economics, some exponents (Dopfer et al., 2004; Dopfer, 2005) have sought to distinguish three 127

Ron L. Martin and Peter J. Sunley

scales of levels of abstraction – those of micro-, meso-, and macro-rules, moving epistemologically from the first to the last.7 Interestingly, in this conceptual schema they suggest that geographical entities such as regions, cities, industrial districts, and the like are examples of meso-scale economics: Work on industrial districts, regional knowledge clusters, learning regions, inter-firm industrial organization, national innovations systems, networks with weak and strong ties, or technical support communities all falls under the heading of meso economics from the evolutionary perspective. (Dopfer et al., 2004, p. 268, fn.9) The potential problem with such a ‘layered’ or ‘stratified’ ontology, however, even if intended to be an overarching or unifying framework, is that it can all too easily lead to the assumption that different theories are applicable at these different levels.8 For economic geographers, however, large-scale historical macro- and meta-processes and structures of capitalist development are not simply ‘out-there’, independent of individual, ‘pre-given’ regions and localities which they (differentially) impact, but are themselves historically constituted by the processes and patterns of ‘regionalisation’ and geographical differentiation of economic activity and relations at any given historical moment. Put another way, national and global processes and structures are in part constituted at the local scale, and the local in part constituted by macro (national and global) processes and structures (see Martin and Sunley, 2012). For geographers, the spatial economy is by its very nature multiscalar: no one ontological level takes precedence, and it is the complex evolving and dialectic relationship between levels that should be the focus of enquiry. What is at issue here, then, is whether an evolutionary ontology is sufficient on its own to provide a theoretical framework capable of explaining the sheer spatio-temporal diversity and complexity of economic development under capitalism.9 Or should attempts be made to integrate or combine it in some way with other perspectives used by economic geographers? In their suggestion for partial theoretical integration, Martin and Sunley (2015) argue that there are good grounds for integrating evolutionary economic geography and geographical political economy, given the latter’s concern with uneven economic development over time. This, they suggest, requires what they call a ‘developmental turn’ in evolutionary economic geography itself (similar to the ‘developmental turn’ that is underway in biological evolutionary theory), so as to align the (mainly micro-level) processes and mechanisms of economic and technological change highlighted in evolutionary economic geography closer to the theorisation of capitalist uneven development found in geographical political economy. A focus on the ‘laws of motion’ of capital accumulation and crisis tendencies under capitalism provides a much-needed ‘deep structure’ system within which micro-level evolutionary processes and mechanisms can be situated and understood. Additionally, institutional ideas are also key: institutional arrangements and regulatory structures both condition the nature of, and are themselves subject to, evolution, conflict, and rupture in the economic landscape. Others, however, prefer to leave geographical political economy out of the mix, because of its critical-normative stance. They argue the focus should be on evolution, institutions, and networks as the main ontological foundations requiring integration on the grounds that these elements figure prominently in all the existing paradigms of economic geography, albeit to different extents (Hassink et al., 2014; Hassink and Gong, 2017). 128

Why an evolutionary economic geography?

Yet others argue that, with its explicit historical imaginary, geographical political economy actually provides a more compelling explanation of spatial economic evolution than does evolutionary economic geography (Mackinnon et al., 2009; Oosterlynk, 2012). However, there is no generally agreed or preferred form of geographical political economy, just as there is no single unified version of evolutionary economics. Nevertheless, most versions of geographical political economy do include an explicit concern with uneven development and an emphasis on large systemic processes (such as regimes of capital accumulation, technology cycles, crises, etc.) and big structures (such as modes of social and institutional regulation, the state, etc.), and the roles these play in shaping the processes, temporalities, and directions of historical and geographical economic change. Evolutionary economic geography and geographical political economy (and also institutionalist economic geography, with its interest in capitalism as an institutionalised socio-cultural formation of formal and informal systems of practices and norms) should not be seen as competing alternative monistic paradigms, but as complementary perspectives each capable of informing the other through a process of dialectical engagement. Economic evolution and economic development should be seen as inextricably intertwined: nothing makes sense in economic geography except in the light of evolution and development.

9.3.3

The importance of addressing key spatial policy issues

Just as evolutionary economics faces a challenge of demonstrating its policy relevance, of translating its theories and empirical findings into meaningful policy recommendations concerning such key issues as (sustainable) growth, innovation, social inequalities, and social welfare, so evolutionary economic geography would benefit from a greater emphasis on addressing spatial economic problems. To be sure, certain of its themes and concepts hold potential in this regard (see, for example, Hassink and Klaerding, 2010). Thus, by demonstrating that economic growth, technological change, and industrial dynamics all have significant geographical dimensions, evolutionary economic geography can help to make and reinforce the case for place-based policy interventions. To date, most of the ‘policy-orientated’ discussions in evolutionary economic geography have focused on the spatial evolution of innovation. Evolutionary economic geography policy work has often tended to marry itself with the regional innovation system approach, as both approaches share similar heterodox roots and a belief that innovation is constrained by system failures. It has been argued that evolutionary economic geography has provided regional innovation systems perspectives with a more analytical and dynamic micro-understanding of the processes behind economic innovation (Isaksen et al., 2018). However, the policy implications and recommendations of evolutionary economic geography have remained largely implicit and under-developed, so that progress has been limited and mixed (Coenen et al., 2017). On the one hand, evolutionary economic geography has carried important policy messages about the need to understand the economic preconditions for regional development and innovation and to foster strategic structural change and industry diversification at a local and regional scale. As Isaksen et al. (2018, page 222) note, ‘This new strategic orientation for regional innovation policies has essentially been informed by evolutionary economic geography, which has offered novel insights into how regional economies transform over time and how new growth paths come into being’. This has rightly emphasised that policies should be place-sensitive and attuned to the characteristics of local economies and that policies should be seen as adaptive experiments which will need to be re-evaluated and changed through time 129

Ron L. Martin and Peter J. Sunley

(Lambooy and Boschma, 2001). On the other hand, however, many of these messages have been highly general and lack practical recommendations. In this gap, smart specialisation policies that argue that regional policy actors should identify new domains based on combinations of old capabilities and assets, by enabling a process of entrepreneurial discovery, have flourished (see Balland et al., 2019). Despite its value in broadening the focus and scope of innovation policy, this approach has also lacked practical guidance on operation and implementation, and its focus on relatedness and related diversification is not unproblematic. As Coenen et al. (2017) argue, it tends to downplay the importance of institutions and agents and treats relatedness as pre-given. As we have argued above, the evidence behind the claims for the benefits of relatedness is by no means as clear and unequivocal as some of the policy pronouncements suggest (Morgan, 2015). Most significantly perhaps, the emphasis on related diversification may well support incremental innovation in established industries, but it is hard to see how it will generate the radical innovations and unrelated diversifications that are required to respond to pressing economic and environmental crises and to revitalise leftbehind peripheral regions (Frenken, 2017; Grillitsch et al., 2018). It is essential then, that evolutionary economic geography continues to develop its own policy thinking and recommendations, and adopts a critical evaluative stance on smart specialisation. Significant scope also exists in relation to the implications for regional and local economic policy of several other themes and concepts that have come to define evolutionary economic geography. For example, an evolutionary perspective could help to inform policy discussions on how best to reconfigure and renew regional economic development paths; on how and why new industries and technologies emerge in some regions and not others; (Coenen et al., 2017; Isaksen et al., 2018); on the variety of possible evolutionary time paths, other than the standard life-cycle trajectory, of business clusters (Martin and Sunley, 2015); on the advantages and disadvantages of regional specialisation versus diversification for long-run regional growth, adaptability, and resilience (Frenken et al., 2007; Martin and Sunley, 2006; Essletzbichler, 2007; Martin et al., 2016); and on the relative advantages of related versus unrelated variety for regional economic success. There is also considerable scope, and need, for studies of how actual spatial policies, spatial governance systems, and policy institutions themselves evolve and co-evolve with regional development outcomes and pathways. In short, evolutionary economic geography has the potential not only to help our understanding of how the spatial economy evolves and changes over historical time, and of the forces of continuity versus transformation involved in this evolution, but by so doing it can help inform and shape the policy interventions needed to ensure that individual regions and places do not get left behind in the historical process of capitalist development. This is not to argue that future research in evolutionary economic geography should in every instance be ‘policy relevant’; it is, however, to suggest that a closer engagement with the challenges confronting spatial policy is not only timely – given the turbulent economic upheavals of our time – but could also positively inform both the empirical remit and the theoretical bases of that research: in other words, what evolutionary economic geography is, how we do it, what we do, and for whom we do it for. Thus far there has been a singular lack of discussion of such normative and axiological issues.

9.4

Conclusion

Arguably the paradigm of evolutionary economic geography has now reached an important juncture in its own evolution. A key question it has to address is whether it is content to 130

Why an evolutionary economic geography?

apply and in some cases adapt evolutionary concepts (and metaphors) borrowed from other (non-spatial) disciplines, or whether (and to what extent) it should strive to construct its own spatial evolutionary ontology, and associated epistemology and theoretical frameworks, based on specifically geographical concepts, principles, and recurring empirical tendencies (such as the various forms and expressions of ‘proximity’, the different types of geographic ‘space’, the relations between agency and systemic conditions; the forces and effects of agglomeration, the dialectic between the ‘space of places’, and the ‘space of flows’, and so on). Some might argue this has already begun to happen, but so far we have taken only small steps in this direction. Perhaps over the coming decade or so, attempts will be made to explore how evolutionary economic geography might be integrated in some way with other perspectives, such as geographical political economy, relational economic geography, and institutionalist economic geography (for suggestions along these lines, see, for example, Martin and Sunley, 2015; Hassink et al., 2014; Hassink and Gong, 2017; Martin, 2020). These different approaches certainly have overlaps and complementarities, and advances on this front seem to us to be possible, and indeed desirable.

Notes 1 As an example, in his otherwise masterly survey of the scope and future research agenda of evolutionary economics, Winter (2017), one of the doyens of evolutionary economics, makes no mention of the role or importance of geography, place or location in the evolutionary process. Nor does Nelson (2020) in his more recent prospectus for evolutionary economics. It is disappointing that these two leading authors, like most other evolutionary economists, make no mention of the work that has been undertaken by economic geographers over the past two decades. 2 Marshall saw these local ‘industrial districts’ as exemplifying what he termed ‘the general rule … that the development of the organism, whether social or physical, involves an increasing subdivision of function between its separate parts on the one hand, and on the other a more intimate connection between them’ (op cit, p.241). This principle (itself a biological analogy) was considered by Marshall to be key to understanding the workings of economic systems (see Martin, 2006; Sunley, 1992). 3 Although, as Reinert and Reinert (2015) show, the idea of ‘creative destruction’ can be traced back earlier to the German economist Werner Sombart and even to the philosopher Friedrich Nietzsche. 4 The literature in evolutionary economic geography is now extensive, and cannot be surveyed in its entirety here. The works just cited contain useful overviews and general surveys of the field and its development over the past decade and a half. 5 For an excellent survey of theories of emergence in complex social systems, see Sawyer (2005) 6 See Sewell (2008) on the temporalities of capitalism. It is interesting, but somewhat curious, that although concerned with time and history, neither evolutionary economists nor evolutionary economic geographers have devoted much attention to problematising the nature of time and temporality (see Martin and Sunley, 2022). Abbott (2001) provides a useful exploration of how time matters in the study of social systems. 7 Others have argued that macro consequences are but the sum of market-driven coordination of micro variety and selection, and that while we can always measure macro-level outcomes, we can only understand economic change as a micro-driven activity ( Metcalfe, 1998; Metcalfe and Ramlogan, 2006). While this is obviously true up to a point, it misses the importance of the notion of emergence and downward causation, as discussed earlier in this chapter. 8 This is what biologists have tended to do, separating questions into the levels of evolutionary origin, current reproductive function, ontogeny, and mechanism. The assumption is that there may be competing theories at any one level, but not between levels – essentially an example of what is sometimes called ‘compatible pluralism’. 9 This is an issue that evolutionary economists could also discuss in relation to their discipline. To argue that an evolutionary ontology is alone sufficient to explain the multifarious dynamics and features of capitalist development is surely to fall victim to the very charge of monism that

131

Ron L. Martin and Peter J. Sunley evolutionary economists levy against mainstream neoclassical economics, which they spend so much time criticising. There is also the interesting fact that in sociology, political science, management, and business studies there has been a growing interest in historical modes of causal investigation to explain social, political, and organisational change and transformation, without recourse to the use of any evolutionary concepts, other than that of path dependence (which concept was developed by an economic historian, again without reference to evolutionary theory).

References Abbott, A. (2001) Time Matters: On Theory and Method, Chicago: University of Chicago Press. Aarstad, J., Kvitastein, O. A. and Jakobsen, S.-E. (2016) Related and unrelated variety as regional drivers of enterprise productivity and innovation: A multilevel study, Research Policy, 45, pp. 844–856. Balland, P.-A., Belso-Martinez, J. and Morrison, A. (2016) The dynamics of technical and business knowledge networks in industrial clusters: Embeddedness, status, or proximity? Economic Geography, 92, pp. 35–60. Balland, P.-A., Boschma, R., Crespo, J and Rigby, D. (2019) Smart specialization policy in the European Union: Relatedness, knowledge complexity and regional diversification, Regional Studies, 53, pp. 1252–1268. Boschma, R. (2004) Competitiveness of regions from an evolutionary perspective, Regional Studies, 38, pp. 1001–1014. Boschma, R. (2015) Towards an evolutionary perspective on regional resilience, Regional Studies, 49, pp. 733–751. Boschma, R. (2017) Relatedness as driver of regional diversification: A research agenda, Regional Studies, 51, pp. 351–364. Boschma R. and Frenken, K. (2011) The emerging empirics of evolutionary economic geography, Journal of Economic Geography, 11, pp. 295–307. Boschma R. and Iammarino, S. (2009) Related variety, trade linkages, and regional growth in Italy, Economic Geography, 85, pp. 289–311. Boschma, R. and Frenken, K. (2018) Evolutionary economic geography, Ch. 11 in Clark, G.L., Feldman, M.P., Gertler, M. and Wójcik, D. (Eds) The New Oxford Handbook of Economic Geography, Oxford: Oxford University Press, pp. 213–229. Boschma, R. and Martin, R. L. (Eds) (2010) Handbook of Evolutionary Economic Geography, Cheltenham: Edward Elgar. Buenstorf G. and Klepper S. (2009) Heritage and agglomeration: The Akron tyre cluster revisited, The Economic Journal, 119, pp. 705–733. Castaldi, C., Frenken, K. and Los, B. (2015) Related variety and unrelated variety and technological breakthroughs: An analysis of US state-level patenting, Regional Studies, 49, pp. 767–781. Chlebna C. and Simmie, J. (2018) New technological path creation and the role of institutions in different geo-political spaces, European Planning Studies, 26, pp. 969–987. Coe, N. (2011) Geographies of production I: An evolutionary revolution? Progress in Human Geography, 35, pp. 81–91. Coenen, L., Moodysson, J. and Martin, H. (2015) Path renewal in old industrial regions: Possibilities and limitations for regional innovation policy, Regional Studies, 49, pp. 850–865. Coenen, L., Asheim, B., Bugge, M. and Herstad, S. (2017) Advancing regional innovation systems: What does evolutionary economic geography bring to the table? Environment and Planning C: Politics and Space, 35, pp. 600–620. Cicerone, G., McCann, P. and Venhorst, V. (2019) Promoting regional growth and innovation: Relatedness, revealed comparative advantage and the product space, Journal of Economic Geography, 20, pp. 293–316. Dawley, S. (2014) Creating new paths? Offshore wind, policy activism and peripheral region development, Economic Geography, 90, pp. 91–112. Dopfer, K., Foster, J. and Potts, J. (2004) Micro-meso-macro, Journal of Evolutionary Economics, 14, pp. 263–279. Dopfer, K. (2005) Evolutionary economics: A theoretical framework, in Dopfer, K. (Ed) The Evolutionary Foundation of Economics, Cambridge: Cambridge University Press, pp. 3–55.

132

Why an evolutionary economic geography? Essletzbichler, J. (2007) Diversity, stability and regional growth in the United States, 1975–2002, in Frenken, F. (Ed) Applied Evolutionary Economics and Economic Geography, Cheltenham: Edward Elgar, pp. 203–229. Essletzbichler, J. (2009) Evolutionary economic geography, insitutions and political economy, Economic Geography, 85, pp. 159–165. Essletzbichler J. and Rigby, D. (2007) Exploring evolutionary economic geographies, Journal of Economic Geography, 7, pp. 549–571. Evans, R. and Karecha, J. (2014) Staying on top: Why is Munich so resilient and successful? European Planning Studies, 22, pp. 205–224. Frenken, K. (2017) A complexity-theoretic perspective on innovation policy, Complexity, Governance and Networks, Special Issue on Complexity and Innovation, 3, pp. 35–47. Frenken, K., Van Oort, F and Verburg, T. (2007) Related variety, unrelated variety and regional economic growth, Regional Studies, 41(5), pp. 685–697. Frenken, K., Cefis, E. and Stam, E. (2015) Industrial dynamics and clusters: A survey, Regional Studies, 49, pp. 10–27. Fritsch, M. and Kublina, S. (2018) Related variety, unrelated variety and regional growth: The role of absorptive capacity and entrepreneurship, Regional Studies, 52, pp. 1360–1371. Gluckler, J. (2007) Economic geography and the evolution of networks, Journal of Economic Geography, 7, pp. 619–634. Grillitsch, M., Asheim, B. and Trippl, M. (2018) Unrelated knowledge combinations: The unexplored potential for regional industrial path development, Cambridge Journal of Regions, Economy and Society, 11, pp. 257–274. Grillitsch, M. and Sotarauta, M. (2020) Trinity of change: Agency, regional development paths and opportunity spaces, Progress in Human Geography, 44, 704–723. Hall, P.G. (1962) The Industries of London Since 1861, London: Routledge. Hassink, R. and Klaerding, C. (2010) Evolutionary approaches to local and regional development policy, in Pike, A., Rodriguez-Pose, A. and Tomaney, J. (Eds) Handbook of Local and Regional Development, London: Routledge, pp. 139–140. Hassink, R., Klaerding, C. and Marques, P. (2014) Advancing evolutionary economic geography by engaged pluralism, Regional Studies, 48, pp. 1295–1307. Hassink, R. and Gong, H. (2017) Sketching the Contours of an Integrative Paradigm of Economic Geography, Paper 2017/12, CIRCLE, University of Lund. Henning, M., Stam, E. and Wenting, R. (2013) Path dependence research in regional economic development: Cacophony or knowledge accumulation? Regional Studies, 47, pp. 1348–1362. Henning, M. (2019) Time should tell (more): Evolutionary economic geography and the challenge of history, Regional Studies, 53, pp. 602–613. Hodgson, G. (2001) How Economics Forgot History: The Problem of Historical Specificity in Social Science, London: Routledge. Isaksen, A. and Trippl, M. (2017) Exogenously led and policy-supported new path development in peripheral regions: Analytical and synthetic routes, Economic Geography, 93, pp. 436–457. Isaksen, A., Tödtling, F. and Trippl, M. (2018) Innovation policies for regional structural change: Combining actor-based and system-based strategies, in A. Isaksen, R. Martin and M. Trippl (Eds) New Avenues for Regional Innovation Systems, Springer, pp. 221–238. Kogler, D. (Ed) (2017) Evolutionary Economic Geography: Theoretical and Empirical Progress, London: Routledge. Lambooy J. and Boschma, R. (2001) Evolutionary economics and regional policy, The Annals of Regional Science, 35, pp.113–131. Levit, G.S., Hossfield, U. and Witt, U. (2011) Can Darwinism be ‘generalised’ and what use would this be? Journal of Evolutionary Economics, 21, pp. 545–562. Mackinnon, D., Cumbers, A., Pike, A., Birch, K. and McMaster, R. (2009) Evolution in economic geography: Institutions, political economy and adaptation, Economic Geography, 85, pp. 129–150. Mackinnon, D., Dawley, Pike, A. and Cumbers, A. (2019) Rethinking path creation: A geographical political economy approach, Economic Geography, 95, pp. 113–135. Marshall, A. (1919) Industry and Trade, London: Macmillan and Company. Martin, R.L. (2006) The localization of industry, Ch. 52 in Raffaelli, T, Beccattini, G. and Dardi, M. (Eds) The Elgar Companion to Alfred Marshall, Cheltenham: Edward Elgar, pp. 393–400.

133

Ron L. Martin and Peter J. Sunley Martin, R. (2010). Roepke lecture in economic geography: Rethinking path dependence: beyond ‘lock‐in’ to evolution. Economic Geography, 86, pp. 1–28. Martin, R.L. (2018) Shocking aspects of regional development: Towards an economic geography of resilience, Ch. 45, in Clark, G.L., Feldmann, M.P., Gertler, M.S. and Wójcik, D. (Eds) The New Oxford Handbook of Economic Geography, Oxford: Oxford University Press, pp. 839–864. Martin, R.L. (2020) Putting the case for a pluralistic economic geography, Journal of Economic Geography, 21, pp. 1–28. Martin, R. L. and Sunley, P. J. (2006) Path dependence and regional economic evolution, Journal of Economic Geography, 6, pp. 395–435. Martin, R. L. and Sunley, P. J. (2007) Complexity thinking and evolutionary economic geography, Journal of Economic Geography, 7, pp. 573–602. Martin, R.L. and Sunley, P.J. (2011) Conceptualising cluster evolution: Beyond the life cycle model? Regional Studies, 45, pp. 1299–1318. Martin, R. L. and Sunley, P. J. (2012) Forms of emergence and the evolution of economic landscapes, Journal of Economic Behaviour and Organisation, 82, pp. 338–352. Martin, R. L. and Sunley, P. J. (2015a) On the concept of regional resilience: Conceptualization and explanation, Journal of Economic Geography, 15, pp. 1–42 Martin, R. L. and Sunley, P. J. (2015b) Towards a developmental turn in evolutionary economic geography? Regional Studies, 49, pp. 712–732. Martin, R.L. and Sunley, P. (2022) Making more of history in evolutionary economic geography, Submitted to The German Journal of Economic Geography. Martin, R. L. and Gardiner, B. (2019) The resilience of cities to economic shocks (and the challenge of Brexit), Papers in Regional Science, 98, pp. 1801–1832. Martin, R.L. and Gardiner, B. (2020) The resilience of Britain’s Core Cities to the Great Recession (with implications for the Covid-19 recessionary shock), in Wink, R. (Ed) Resilience for Regions, Organisations and Management, Cheltenham: Edward Elgar. Martin, R. Sunley, P. Gardiner, B. and Tyler, P. (2016) How regions react to recessions: Resilience and the role of economic structure, Regional Studies, 50, pp. 561–585. Martin, R., Sunley, P., Gardiner, B. Evenhuis, E. and Tyler, P. (2018) The city dimension of the productivity growth puzzle: The relative role of structural change and within-sector slowdown, Journal of Economic Geography, 18, pp. 539–570. Maskell, P. and Malmberg, A. (2007) Myopia, knowledge development and cluster evolution, Journal of Economic Geography, 7, pp. 603–618. Massey, D. (1992) Politics and space/time, New Left Review, 196, pp. 65–84. Metcalfe, J.S. (1998) Evolutionary Economics and Creative Destruction, London: Routledge. Menzel, M. and Fornahl, D. (2010) Cluster life-cycles – dimensions and rationales of cluster evolution, Industrial and Corporate Change, 19, pp. 205–238. Metcalfe, J.S., Foster, J. and Ramlogan, R. (2006) Adaptive economic growth, Cambridge Journal of Economics, 30, pp. 7–32. Morgan, K. (2015) Smart specialisation: Opportunities and challenges for regional innovation policy, Regional Studies, 49, 480–482. Myrdal, G. (1957) Economic Theory and Underdeveloped Regions, London: Gerald Duckworth. Neffke, F., Henning, M. and Boschma, R. (2011) How do regions diversify over time? Industry relatedness and the development of new growth paths in regions, Economic Geography, 87, pp. 237–265. Nelson, R. (2020) A perspective on the evolution of evolutionary economics, Industrial and Corporate Change, 29, pp. 1101–1119. Oosterlynk, S. (2012) Path dependence: A political economy perspective, International Journal of Urban and Regional Research, 36, pp. 158–165. Reinert, H. and Reinert, E.S. (2015) Creative destruction in economics: Nietzsche, Sombart, Schumpeter, New Nietzsche Studies, 9, Issue 3/4, pp. 1–23. Rigby, D. (2018) Evolutionary economic geography: An emerging field or framework? Chapter 8 in Cook, G., Johns, J., MacDonald, F., Beaverstock, J. and Pandit, N. (Eds) The Routledge Companion to the Geography of International Business, London: Routledge. pp. 129–146. Rigby, D. and Brown, W.M. (2015) Who benefits from agglomeration? Regional Studies, 49, pp. 28–43. Sensier, M., Bristow, C. and Healey, A. (2016) Measuring regional economic resilience across Europe: Operationalising a complex concept, Spatial Economic Analysis, 11, pp. 128–151.

134

Why an evolutionary economic geography? Sewell, W. H. (2008) The temporalities of capitalism, Socio-Economic Review, 6, pp. 517–537. Simmie, J. and Martin, R. L. (2010) The economic resilience of regions: Towards an evolutionary approach, Cambridge Journal of Regions, Economy and Society, 3, pp. 27–43. Stam, E. (2009) Why butterflies don’t leave: Locational behaviour of entrepreneurial firms, Economic Geography, 83, pp. 27–50. Sawyer, R.K. (2005) Social Emergence: Societies as Complex Systems, Cambridge: Cambridge University Press. Stoelhorst, J.W. (2014) The future of evolutionary economics is in a vision from the past, Journal of Institutional Economics, 10, pp. 665–682. Storper, M. (2009) Roepke lecture in economic geography: Regional context and global trade, Economic Geography, 85, pp. 1–21. Stam, E. (2015) Entrepreneurial ecosystems and regional policy: A sympathetic critique, European Planning Studies, 23, pp. 1759–1769. Strambach, S. (2010) Path dependence and path plasticity: The co-evolution of institutions and innovation – the German customized business software industry, in Boschma, R. and Martin, R. L. (Eds) Handbook of Evolutionary Economic Geography, Cheltenham: Edward Elgar, pp. 406–431. Sunley, P.J. (1992) Marshallian industrial districts: The case of the Lancashire cotton industry in the inter-war years, Transactions Institute British Geographers NS, 17, 306–320. Storper, M., Kemeny, T., Makarem, N. and Osman, T. (2015) The Rise and Fall of Urban Economies: Lessons from San Francisco and Los Angeles, Stanford: Stanford University Press. Trippl, M., Grillitsch, M, Isaksen, A. and Sinozic, T. (2015) Perspectives on cluster evolution: Critical review and future research issues, European Planning Studies, 23, pp. 2028–2044. Trippl, M., Grillitsch, M. and Isaksen, A. (2018) Exogenous sources of regional industrial change: Attraction and absorption of non-local knowledge for new path development, Progress in Human Geography, 42, pp. 687–705. Wink, R., Kirchner, L., Koch, F. and Speda, D. (2016) There are many roads to reindustrialisation and resilience: Place-based approaches to three German regions, European Planning Studies, 24, pp. 463–488. Winter, S. (2017) Pursuing the evolutionary agenda in economics and management research, Cambridge Journal of Economics, 41, pp. 721–747.

135

10 DARWIN’S IDEAS AND THEIR MIXED RECEPTION IN EVOLUTIONARY ECONOMICS 1 Gabriel Yoguel and Verónica Robert

10.1

Introduction

Since the appearance of Nelson and Winter’s An Evolutionary Theory of Economic Change (1982), those interested in evolutionary economics have debated the role that Darwin’s ideas may play in explaining economic dynamics. However, evolutionary economics did not start in the 1980s. Several authors throughout the history of economic thought resorted to biology and biological evolution to explain economic and social phenomena (Marshall, 1920; Schumpeter, 2003 and 2013; GeorgescuRoegen, 1970; and Boulding, 1978; among others). American institutionalist authors like Thorstein Veblen and John Commons 1889; 1934) particularly recognized Darwin’s transcendental contribution to different branches of science and the deeper implications of this for philosophy and politics (Hodgson, 2004). In modern evolutionary economics, biological references are usually accompanied by justifications regarding the possibility of sharing rules between the biological and socioeconomic realms. These concerns have raised some questions about the heuristics and epistemological assumptions of evolutionary economics (Robert, Yoguel, and Llerena, 2017), notably regarding whether the economy shares the same building blocks as biology. There is also an epistemological question: is using Darwin’s ideas and evolutionary principles conducive to explaining economic change, even though economy and biology are not made of the same building blocks? Between 1995 and 2015, a set of articles appeared addressing the issue of applying Darwin’s concepts to economics (Hodgson, 2002 and 2007; Hodgson and Knudsen, 2006; Dopfer and Potts, 2004, Stoelhorst, 2008, Aldrich et al., 2008; Witt, 2006; Witt and Cordes, 2007; Cordes, 2007; Foster, 1997 and 2005; Nelson, 2006; Nelson and Winter, 2002). In this chapter, we organize the contributions to the debate into four positions, some of which have already been identified in the literature (Witt, 2006; Vromen, 2008). The first category considers Darwin’s ideas only in the field of metaphors and analogies (MA) (Metcalfe, 2002; Nelson and Winter, 1982). The second, generalized Darwinism (GD) (Hodgson, 2007; Hodgson and Knudsen, 2006), departing from Dawkins’s notion of universal Darwinism posits that any evolutionary process follows the Darwinian principles of

136

DOI: 10.4324/9780429398971-12

Darwin’s ideas and their mixed reception in evolutionary economics

variation, selection, and retention. The third postulates a continuity hypothesis (CH), which stresses a connection between biological, cultural, and economic evolution, although with rules that differ. The final position is the most distant from Darwin: the introduction of complexity theory (CT) allows us to look at the formal structures underlying any emerging process of change (Foster 1997. CT includes a more comprehensive formalization of an evolutionary process (Dopfer and Potts, 2004), guided by self-organization. A general definition of the economic system as a complex, adaptive, self-organized system is provided in which the dynamics are an open-ended process (Dopfer and Potts, 2004; Hodgson, 2004). This definition comfortably accommodates different branches of evolutionary economics, without mentioning Darwin. Over time, this debate has become rather inconclusive, so, in this chapter, we seek to see if this is because of the different ways that Darwin has been used to explain economic dynamics and innovation. The main question is whether staying in the field of metaphors and analogies or abandoning Darwinian evolutionary thinking, either in essence (GD) or as a continuity (CH), constrains or enhances the explanatory power of modern evolutionary economics. In sum, we wish to explore the role that Darwinian ideas play in building sense and meaning within the innovation process in a modern economy as a theory, metatheory, or epistemology. The chapter is structured as follows. Section 2 explains the four uses of Darwin’s ideas in evolutionary economics. Specifically, section 2.1 defines the analytical perspective of MA, section 2.2 discusses the position of GD, section 2.3 presents the CH, and section 2.4 discusses CT as a way to expand beyond Darwin’s ideas. In section 3, we present the debate between the four positions and analyze the extent to which the different uses of Darwin’s ideas found in each group allow us to better understand evolutionary economic processes. Finally, section 4 contains our main conclusions.

10.2

Four positions regarding the relevance of Darwin’s ideas for evolutionary economics 10.2.1

Metaphors and analogies

The first position states that Darwinian ideas can be used only in the terrain of MA, since the biological realm is intrinsically different to the social one (Metcalfe, 1998; Nelson and Winter, 1982, 2002; Nelson, 2006). This position is probably the most accepted one in evolutionary thinking. It rests on the premise that Darwinian evolutionary theory may act as a source of inspiration but argues that it is not possible to claim that the economy is affected by the same rules as biology. MA authors draw on to ideas taken from Darwinian evolution, going to great lengths to clarify that they are not applying them “as if” economy behaves as biology does. For instance, Metcalfe (1998) states that although evolution is a central concept in biology, its use does not mean resorting to biological explanations. As a consequence, variation, retention, and selection are not general principles but rather allow us to imagine how evolutionary processes can happen in economic realms, meaning generation and resolution of economic diversity (Metcalfe, Foster and Ramlogan, 2006). The explanatory power of evolutionary theories about endogenously changing systems has led the authors in this group to argue that evolutionary theorization can be key to the foundation of a heterodox theory of innovation and technological change. Resorting to evolutionary arguments results in new analytical tools (such as evolutionary models like replicator dynamics, derived from Price) that endow evolutionary theories with the formal rigor that the discipline is seen as requiring (Nelson and Winter, 2002).

137

Gabriel Yoguel and Verónica Robert

However, Nelson (2006) establishes limits to the application of biological ideas to economic evolution. According to the author, a narrow form of universal Darwinism (see next section) should not be acceptable to social scientists, since the details of cultural evolution are incompatible with biological evolution. He also warns that social scientists in different fields could be tempted to unify dynamics under the umbrella of variation, selection, and retention, suppressing the variety of forms of evolution. Instead, following Hull (1973) and Boyd and Richerson (2005), he proposed a broader form of universal Darwinism that is compatible with the building of a new evolutionary theory based on analogies. This form is also open to significant differences between the way aspects of human culture evolve and the evolution of the species (Nelson, 2006). Taking Darwin’s ideas through metaphors and analogies allows evolutionary economics to undertake dynamic analyses of economic change at the organization and industry levels, particularly in contexts where innovation and technological change are the main drivers of competition2. The focus on competition led to other issues such as returns and profiting from innovation, entrepreneurial activity, distribution of firm size and market structure determinants, learning, and competence-building. All of these issues constitute the strengths of the evolutionary approach, due to the emphasis on learning and immanent novelty that derive from the complex interactions among organizations that cannot be rationally foreseen though they can be explained afterward.

10.2.2

Generalized darwinism

The GD position suggests that the evolutionary, open, dynamic, complex systems used to explain economic evolution follow Darwinian principles of variation, selection, and retention (Hodgson and Knudsen, 2006; Aldrich et al., 2008; and Stoelhorst, 2008). This position, which is based on Dawkins’s (1983) and Dennett’s (1996) concept of universal Darwinism, postulates that any evolutionary process (be it biology, technology, linguistics, or economics) can be understood in terms of a process where the key principles are neatly analogous to those described by Darwin in biological evolution. They consider that these three principles constitute the basic ontological elements of any evolutionary theory seeking to explain immanent change (Hodgson and Knudsen, 2006). The authors holding this position define Darwinian principles abstractly, devoid of any biological content, as follows: (i) the creation of variety through the introduction of novelty, (ii) interaction between the individual and the environment that proves novelty (entity subject of selection), and (iii) the retention of attributes that fit better with the environment, generating irreversible changes at the population level. According to them, evolutionary ontology is contained in these three abstract principles, since the way in which they explain population change is nondomain-specific. These authors support Hull’s (1973) thesis that Darwinian principles provide a general explanatory framework into which particular explanations and empirical details have to be placed, since the three principles will assume different forms in different evolutionary domains. Darwinian general principles are like empty vases. They may contain several fieldspecific features that apply only to economic, cultural, or technological evolution. To complete the ontology of evolutionary economics, it is necessary to define the specific features of the three evolutionary principles. Authors including Stoelhorst (2008) and Aldrich et al. (2008), along with other authors, developed this research program specifying entities of selection (i.e., routines, customs, institutions) and the specific mechanisms 138

Darwin’s ideas and their mixed reception in evolutionary economics

of variation (learning), selection (interactions), and retention (imitation, firm growth) by which firms adapt to their environment. On the topic of variation, the generation of biological variety requires genetic recombination and mutations, but in the social sciences, it requires intentional and idiosyncratic behavior in an uncertain environment: innovation, imitation, learning, interactions, and planning are elements of those behaviors that feed the creation of variety. In turn, the selection process happens through the market, although it is considered as a social construction that conditions the way it operates by social technologies, including norms, rules, standards, and differential power of competing agents. Selection will determine the growth of firms as well as the spread of technologies and best practices. At the same time, the interaction networks among organizations open up channels for knowledge circulation and help explain why some technologies only spread locally, based on the notion that different absorption capacities account for idiosyncratic adoption and adaptation to specific contexts. As Hodgson (2002) stresses, the capacity for (i) self-reflection, (ii) reasoning, (iii) deliberative behavior, and (iv) forecasting and planning activities are characteristics of human beings in the economic realm. Consequently, he argues that Darwinian evolution cannot be blind when it is applied to different domains. Therefore, he stresses that Darwinism applies fully to socioeconomic systems and involves: i) a general theory of the evolution of all open complex systems; ii) a basic philosophical commitment to detailed, cumulative, causal explanation; and iii) human intentionality. Another issue brought up by Hodgson (2004) is the American institutionalist position (Dewey) regarding uncaused causes. This is taken from Charles Darwin (1859), who assumes that science should look at the final cause of each phenomenon rather than embracing a notion of a spontaneous, uncaused event. Therefore, all intentions, behaviors, and institutions have to be explained by a causal process. Human behavior is part of social reality and social interactions involve human expectations concerning the intentions of others. None of these points is undermined by the recognition that intentions themselves are caused (Hodgson, 2004).

10.2.3

The continuity hypothesis

The CH position (Witt 2006, 2006; Witt and Cordes, 2007; Cordes, 2007, 2007) states that there is a connection between natural biological evolution and economic evolution. According to this hypothesis, natural selection established the foundations for evolutionary processes such as those occurring in social and economic realms. In this regard, social evolution relies on a bedrock shaped by biological evolution. In this context, it should not be necessary to resort to analogies and metaphors (MA), nor to assume that the same Darwinian evolutionary principles apply to the social realm (GD). Instead, specific principles of social, cultural, and economic evolution should be established derived from previous stages in human history. CH starts by considering biological and social evolution as distinctive phenomena whose starting point could be said to be the “great leap,” when Homo Sapiens began social life, around 50,000 years ago (Taniguchi, 2009). At this moment, cultural and social evolution drifted from biological evolution. The natural selection of traits was left behind by the intentional interaction of humans with the environment. Natural evolution defines the limits of the subsequent evolution of culture, society, and the economy. In these realms, natural selection does not apply but is instead replaced by a 139

Gabriel Yoguel and Verónica Robert

cultural, social, economic process. As CH authors state, Darwinian selection can explain the origins of cognitive abilities (human learning and intentional and deliberative behaviors) but cannot account for variations in capacities that explain the current dynamics of cultural and economic evolution. They state that economic evolution has its own rules, which should be investigated by the research program of evolutionary economics. According to CH, genetics conditioned human skills at the beginning of culture but did not affect the subsequent stages of cultural evolution. The historical processes of economic evolution, which are part of cultural evolution, are embedded within limits established by the natural evolution operating in previous stages. Darwinian theory thus explains the origins of economic evolution in human phylogeny and fosters understandings of the lasting influence of natural elements, dispositions, and programs on human behaviors. Cordes (2007) also considered that “the biological foundation of learning and reasoning allow and directly affect cultural evolution.” Evolutionary selection has created a set of principles that play a part in generating human behavior. These repertoires of behavior are the basis on which other forms of evolution rest. Cordes (2007) highlights the centrality of this in building a social evolutionary ontology. In sum, according to the CH position, the ontological continuity between biology and cultural evolution does not deny the profound differences between the biological realm and the cultural and social realm but nonetheless establishes points of contact between them.

10.2.4

Complexity theory as an improvement of Darwin’s ideas

The fourth position explains the innovation process and evolutionary dynamics using complex systems theory.3 In this regard, it draws on a more general framework than Darwinian evolution to explain economic change, which included the process of self-organization (Foster, 2005) and cumulative causation (Dosi, 2014; Saviotti and Pyka, 2004), innovation as a change in routines (Nelson and Winter, 1982), and the dynamics of the system. From this perspective, the heterogeneous development of capacities and connections between system components generates continuous changes in the structure, which emanates from microdynamics. This is defined in terms of the different economic performance among firms, as well as differences in the possibility of profiting from innovation and the appropriation of rents in the competition process. In this vein, some authors (Dosi and Teece, 1998) hold that the selection process could occur both among firms and within firms, through management selecting best practices, which could lead to better appropriation of scale economies and concentration of capital. Authors who take the CT position avoid making references to Darwin, since they argue that complex systems theory contains a broader formal framework for explaining economic change without resorting to biological references as a metaphor, generalization, or continuity. Foster (1997:444) emphasizes that “once biological analogy is replaced in favor of a selforganization perspective, there is no longer any interest in the microscopic details of selection mechanisms, but in the endogenous tendency for acquired knowledge and skills to interact and to create increases in economic organization and complexity.” However, a self-organization perspective does not imply setting Darwinian ideas aside. Evolutionary biology adopted self-organization as a source of variety resolution that complements Darwinian competition. In the same vein, evolutionary economics also can enlarge the basis of its understanding by combining self-organization with competition (Foster, 2005). 140

Darwin’s ideas and their mixed reception in evolutionary economics

From this approach, emergent properties are key to understanding the dynamics of organization and structural change. Emergent properties lead to self-organization processes, which are required to accelerate systemic change (Kaufman, 1995; Foster, 2005, Metcalfe, 1998). Complexity allows us to understand the effects on the micro, meso, and macrostructure of heterogeneous and decentralized agent behaviors. According to Foster (2005), population competition is not enough to explain change since it may emerge from inside firms that anticipate the selection process (Dosi and Teece, 1998). In this vein, creativity and cooperation among agents are characteristic of capitalistic dynamics, along with the struggle associated with the competition process. According to Foster, the second law of thermodynamics and dissipative structures make it possible for us to engage in an evolutionary analysis that includes historical trends. From this perspective, economic evolution is a process of cumulative, nonlinear irreversible structural change that involves the acquisition of energy and knowledge and yields creativity in economic evolution. In this regard, other authors (Antonelli, 2011; Arthur, 2009) stress the relevance of systemic perspectives and cumulative causation change, wherein positive feedback, network externalities, and emergent properties do not draw on any Darwinian perspectives. Saviotti and Pyka (2004) place the cumulative causation process at the center of their explanation of structural change, which is considered to be an emergent property of a complex system. They do not quote Darwin to explain dynamic structural change. Nor does Edquist (1997) resort to Darwin when he observes that the creation of microeconomic capabilities relies upon institutional context. In sum, the complexity perspective allows us to understand the generation of variety and innovation as: i) the result of problem-solving process in existing routines (Nelson and Winter, 1982), ii) mechanisms of self-organization (Foster, 1997), iii) the development of capacities in and interactions between system components within an institutional context (Antonelli, 2014; Edquist 1997), iv) mechanisms of cumulative causation (Arthur, 2009; Saviotti and Pyka, 2004) and v) selection occurring within the firm (Dosi and Teece, 1998). In this way, within CT, the phenomena of variety and the emergence of innovation are mostly independent of Darwinian thought.

10.3

The heuristics and epistemological principles implied in each position

Having defining the four positions (2.1 to 2.4) and their relationship with Darwin’s ideas, in this section, we present the main argument of the debate. We will also explore how far each of the four positions accomplishes the task of providing a useful methodological strategy for evolutionary economics. First, as was suggested before, MA, GD, and CH are closer to Darwin than CT, which is based on a general framework that could be considered broader than that of Darwin. However, the ways in which MA, GD, and CH draw on Darwin’s ideas are completely different: (i) MA stresses a metaphorical use of these ideas; (ii) GD directly applies an abstract form of the Darwinian notions of variation, selection, and retention to explain economic evolution, and (iii) CH sustains an ontological and historical continuity between biological and economic evolution. However, the four positions do agree that the economy is governed by different rules than biology, pointing at the specific cognitive capacities, learning processes, and creativity that individuals have in the cultural and economic context and how these capabilities interact with selection mechanisms. 141

Gabriel Yoguel and Verónica Robert

Second, relative to a long-term historical perspective, CH authors stress that it is necessary to research the specific forms that evolution takes at each historical stage and their links with previous stages. In this sense, CH states that current cultural and economic evolution is an emergent property of the previous evolutionary process. That is, evolution regenerates its own evolutionary rules. In this sense, CH’s perspective is naturalistic as long as it holds that current evolutionary conditions are derived from past evolutionary conditions. In contrast, MA and GD, in particular, are only interested in current socioeconomic evolution and describing the general principles that govern it. MA sees Darwin as a source of conceptual inspiration but distances itself from it when it comes to applying it to the economic field. In turn, GD states the existence of fundamental principles that are common to any process of change, with the main goal of building a theory that avoids uncaused causes. By resorting to an abstract structure of evolutionary principles, GD abandons a long-term historical explanation, which links current behavior to cognitive capacities and linkages built up over tens of thousands of years of human evolution. Third, GD critiques the MA position when emphasizing that resorting to metaphors and analogies avoids making the variation, selection, and retention principles explicit since it does not identify either the unit of selection or the sources of variation and retention (Stoelhorst, 2008; Hodgson, 2004). It is argued that GD overcame the ambiguities of MA by identifying the domain-specific contents for the principles of variation, selection, and retention. According to DG, Darwinism—especially the selection process—is a theory that needs to be taken into account and cannot be replaced by self-organization. The theory of self-organization applied to microevolution can prove useful for explaining how individual entities develop by themselves, but Darwinian thinking is needed to explain the evolution of the whole system since it focuses on population thinking. However, GD alone does not provide a full detailed explanation of evolutionary processes or outcomes. It is a metatheoretical framework in which Darwinian general principles are needed to provide an evolutionary theory of social science but are insufficient in themselves. Fourth, in contrast to MA, DG, and CH, CT is oriented toward identifying the principles of structural change, self-organization, and cumulative causation that underlie complex systems. That is, emergent properties place a key role in explaining evolutionary and innovative dynamics. Emergent properties at different levels of analysis may explain historical change without the need to define specific forms of variation, selection, and retention or evolutionary entities, which in the view of some authors constitutes the failure of DG (Foster, 2005). Therefore, it is not Darwinian principles that are at stake, but broader and more general rules. In this sense, CT distances itself from the possibility of applying Darwin’s principles, although in a different way to MA or CH. According to Foster (2005), while creativity and cooperation—not competition—are the main heuristics and ontological drivers of structural change and innovation in self-organization, the heuristics and ontology of DG predominate in the idea of competitive struggle. CT also includes cumulative causation, positive feedback, and increasing returns, which helps explain global system dynamics. Up to here, we have presented the debate among the four different positions regarding the possible use of Darwinian thought in evolutionary economics. We show that there are different assumptions made by each of the four positions that lead to different epistemological choices. Although there are several possibilities regarding epistemological perspectives on evolutionary economics that can be found in the work of different authors, there are three basic 142

Darwin’s ideas and their mixed reception in evolutionary economics

epistemological principles that are common to most of them: (i) realism, (ii) open-ended dynamics, and (iii) nonreductionism (Robert et al., 2017). In terms of realism, evolutionary economics states that the process of building the theory must begin with an “appreciative vision” (Nelson, 1991) and requires the rejection of an ad-hoc axiomatic formulation on which the theory is built. This does not imply a rejection of any possibility of formalization but rather of those that do not dialogue continuously with an appreciative theorization, in which the commitment to realism is greater. The open-ended dynamics principle assumes that the economic system is embedded in a process that occurs historically and in real-time, characterized by radical uncertainty, path dependence, and irreversibility, such that predictions are not feasible and explanations should be based on the description of a dynamic process. Finally, the principle of nonreductionism refers to the fact that the aggregate is different from the linear sum of its parts. Individual actions and behaviors have macro effects that are difficult to predict, since multiple micro-macro interactions take place along the system’s dynamics (Dopfer, 2004). These three principles are present within a wide range of conceptualizations in neoSchumpeterian evolutionary perspectives, even though there are significant differences within this group (Robert, Yoguel, & Lerena, 2017). The four positions on Darwin’s ideas might act as metatheories that support the conceptual developments of neo-Schumpeterian evolutionist perspectives. Therefore, they should comply with all three of these epistemological principles. However, this is actually only true for two of the three principles: nonreducibility and open-ended dynamics. However, this is not so evident in the case of realism. The level of abstraction managed by GD authors does not allow us to see how the epistemological premise of realism works. For example, Foster (2005) argues that GD does not allow us to build empirical analyses based on historical processes. Nor does CH fully comply with realism as a fundamental epistemological premise. The explanation of cultural and economic evolution as an emergent cause of earlier evolutionary processes breaks with the idea of using empirical data as the primary source of information for analysis. The continuity hypothesis between evolutionary processes is difficult to addressed empirically, since qualitative changes in evolution arise as an emergent property of prior evolutionary forms in long-term history. The continuity hypothesis is thus not constructed to be empirically verified, nor does it emanate from a set of observations and characterizations of the evolutionary form of contemporary economic and social processes. In contrast, it is a theorization that contributes to a unified theory between different fields yet is not evident in its manifestation: CH aims to unveil very long-term processes, which are unobservable. Therefore, among the four positions, we find that MA and CT are better suited to grasping economic evolutionary dynamics. Specifically, these two positions are constructivist, in terms of the quest for a realistic theory that takes feedback among formal and narrative theorizing into account, based on observation and abstractions. MA provides a realist epistemology based on stylized facts, and CT is congruent with these premises when look forward replicating attempting to replicate the historical process. This is the case of the history-friendly models among other many evolutionary models ranging from the Price equation (Metcalfe, 1998) to evolutionary agent-based models. Consequently, neither MA nor CT can be discarded. CT is broad enough to address multiple fields of study with formal tools that are also compatible with historical perspectives. 143

Gabriel Yoguel and Verónica Robert

MA provides a framework to appreciative theory that cannot be disregarded when approaching an ever-changing phenomenon, one in which observation and theorizing require an exercise of imagination that no formal framework can provide.

10.4

Conclusions

In this chapter, we have analyzed four positions regarding Darwin’s ideas. We have outlined the differences in their epistemological principles. The four positions agree on the value of a quest for a general framework that explains evolution and change. In searching for a theory of change, each of them has to some extent established a dialogue with classical and modern biological evolution and complexity theory: elements like interacting microcomponents and the macrostability that emerges from them can be found in all four positions-. Although these attempts did not end in a unified theory, they could be understood as a search for a metatheory. The four positions establish that the specificities of the economic realm exceed solely biology-based explanation and require additional theoretical and methodological developments. The four positions assume that different paths can be followed to achieve this: (i) a path that is absolutely independent of Darwin’s ideas (MA), (ii) one that is treated as a particular case of Darwinian theory (DG), (iii) one that is studied as an emergent property of Darwinian evolution (CH), or (iv) one that is conceptualized as a nested theory within a general framework of complexity systems (CT). In this sense, the debate among the four positions poses an epistemological dispute that has not yet been solved. However, this disagreement has brought about some progress toward the development of a common metatheoretical framework for understanding economic evolution. It seems that Darwin’s contribution should not be judged by how effectively it can be applied in the field of economic evolution but by whether it can provide a background for the conceptual discussion. Once we have recognized that the four positions entail heuristics and methodologies that differ from each other, some questions remain unanswered. Has one position been more successful than the other three in providing a useful epistemological and methodological strategy? Can CT contain the other three under a single umbrella or does it make more sense to assume that each position will play a particular role in different research questions? In that case, would an epistemological fragmentation be functional for evolutionary economics, or should we attempt to arrive at common epistemological principles?

Notes 1 Our thanks to Kurt Dopfer and John Foster for their comments on an earlier version of this chapter. 2 The competition process requires the presence of social technologies ( Nelson and Sampat, 2001) that constitute a necessary condition for incorporating physical technologies. 3 A complex system is made up of a large number of parts that interact in a way that makes it difficult to infer the properties of the whole by knowing the properties of the parts and the laws of their interaction ( Simon, 1962). A system is complex if it is modular (decomposable into several interacting subsystems), open (made up of components that change over time in exchanges of information and energy with the environment) and hierarchical (each subsystem is also a complex system) ( Dopfer et al., 2004; Foster, 2005; Metcalfe and Foster, 2007).

144

Darwin’s ideas and their mixed reception in evolutionary economics

References Aldrich, H.E., Hodgson, G.M., Hull, D.L., Knudsen, T., Mokyr, J., & Vanberg, V.J. (2008). In defence of generalized Darwinism. Journal of Evolutionary Economics, 18(5), 577–596. 10.1007/s001 91-008-0110-z Antonelli, C. (2011). Handbook on the Economic Complexity of Technological Change. Edward Elgar Publishing. Arthur W.B. (2009). The Nature of Technology: What it is and How it Evolves. Simon and Schuster. Boulding, K.E. (1978). Ecodynamics: A New Theory of Societal Evolution. SAGE Publications, Incorporated. Boyd, R., & Richerson, P.J. (2005). The Origin and Evolution of Cultures. Oxford University Press. Cordes, C. (2007). Turning economics into an evolutionary science: Veblen, the selection metaphor, and analogical thinking. Journal of Economic Issues, 41(1), 135–154. 10.1080/00213624.2007.115 06998 Dennett, D. C. (1996). Darwin’s Dangerous Idea: Evolution and the Meanings of Life. Simon and Schuster. Dopfer, K. (2004). The economic agent as rule maker and rule user: Homo Sapiens Oeconomicus. Journal of Evolutionary Economics, 14(2), 177–195. 10.1007/s00191-004-0189-9 Dopfer, K., Foster, J., & Potts, J. (2004). Micro-meso-macro. Journal of Evolutionary Economics, 14(3), 263–279. 10.1007/s00191-004-0193-0 Dopfer, K., & Potts, J. (2004). Evolutionary realism: A new ontology for economics. Journal of Economic Methodology, 11(2), 195–212. 10.1080/13501780410001694127 Dosi, G., & Nelson, R.R. (1994). An introduction to evolutionary theories in economics. Journal of Evolutionary Economics, 4(3), 153–172. 10.1007/BF01236366 Dosi, G., & Teece, D.J. (1998). Organizational competencies and the boundaries of the firm. In: Arena, R., & Longhi, C. (eds) Markets and Organization. Springer, Berlin, Heidelberg. 10.1007/ 978-3-642-72043-7_12 Dosi, G. (2014). Dinámica y coordinación económica. Algunos elementos para un paradigma alternativo evolucionista. In: Barletta, F., Robert, V., & Yoguel, G. (eds.) Tópicos de la teoría evolucionista neoschumpeteriana de la innovación y el cambio tecnológico. Polvorines: Universidad Nacional de General Sarmiento, Ciudad Autónoma de Buenos Aires: Miño y Dávila. Edquist, C. (1997). Systems of innovation approaches–their emergence and characteristics. Systems of Innovation: Technologies, Institutions and Organizations, 1989, 1–35. Foster, J. (1997). The analytical foundations of evolutionary economics: From biological analogy to economic self-organization. Structural Change and Economic Dynamics, 8(4), 427–451. 10.1016/ S0954-349X(97)00002-7 Foster, J. (2005). From simplistic to complex systems in economics. Cambridge Journal of Economics, 29(6), 873–892. 10.1093/cje/bei083 Georgescu-Roegen, N. (1970). The economics of production. The American Economic Review, 60(2), 1–9. https://www.jstor.org/stable/1815777 Hodgson, G.M. (2002). Darwinism in economics: From analogy to ontology. Journal of Evolutionary Economics, 12(3), 259–281. 10.1007/s00191-002-0118-8 Hodgson, G.M. (2004). The Evolution of Institutional Economics. Routledge. Hodgson, G.M. (2004). Darwinism, causality and the social sciences. Journal of economic methodology, 11(2), 175–194. Hodgson, G.M. (2007). The revival of Veblenian institutional economics. Journal of Economic Issues, 41(2), 324–340. 10.1080/00213624.2007.11507019 Hodgson, G.M., & Knudsen, T. (2006). The nature and units of social selection. Journal of Evolutionary Economics, 16(5), 477–489. 10.1007/s00191-006-0024-6 Hull, D.L. (1973). Darwin and his critics: The reception of Darwin’s theory of evolution by the scientific community. Harvard Univ. Press. Kaufman, S. (1995). At home in the universe: The search for laws of complexity. Marshall, A. (1920). Principles of Economics. Book VI. London. 618–619. Metcalfe J.S. (2002). Evolutionary Economics and Creative Destruction. Routledge. Metcalfe, J. S., Foster, J., & Ramlogan, R. (2006). Adaptive economic growth. Cambridge Journal of Economics, 30(1), 7–32. 10.1093/cje/bei055 Metcalfe, J. S., & Foster, J. (2007). Evolution and Economic Complexity. Edward Elgar Publishing.

145

Gabriel Yoguel and Verónica Robert Nelson, R. (2006). Evolutionary social science and universal Darwinism. Journal of Evolutionary Economics, 16(5), 491–510. 10.1007/s00191-006-0025-5 Nelson, R.R. (1991). Why do firms differ, and how does it matter? Strategic Management Journal, 12(S2), 61–74. Nelson, R.R., & Winter, S.G. (1982). An Evolutionary Theory of Economic Change. Harvard University Press. Nelson, R.R., & Sampat, B.N. (2001). Making sense of institutions as a factor shaping economic performance. Revista de Economía Institucional, 3(5), 17–51. http://www.scielo.org.co/scielo.php? script=sci_abstract&pid=S0124-59962001000200002&lng=en&nrm=iso&tlng=es Nelson, R.R., & Winter, S.G. (2002). Evolutionary theorizing in economics. Journal of Economic Perspectives, 16(2), 23–46. 10.1257/0895330027247 Robert, V., Yoguel, G., & Lerena, O. (2017). The ontology of complexity and the neo-Schumpeterian evolutionary theory of economic change. Journal of Evolutionary Economics, 27(4), 761–793. 10.1007/ s00191-017-0512-x Saviotti, P.P., & Pyka, A. (2004). Economic development by the creation of new sectors. Journal of Evolutionary Economics, 14(1), 1–35. 10.1007/s00191-003-0179-3 Schumpeter, J. (2003). The Theory of Economic Development (J.Backhaus, Ed.). Springer US. 10.1007/ 0-306-48082-4_3 Schumpeter, J.A. (2013). Capitalism, Socialism and Democracy. Routledge. Simon H.A. (1962): The Architecture of Complexity, Proceedings of the American Philosophical Society, 106, 467–482. Stoelhorst, J.W. (2008). The explanatory logic and ontological commitments of generalized Darwinism. Journal of Economic Methodology, 15(4), 343–363. 10.1080/13501780802506661 Taniguchi, K. (2009). A thought on the continuity hypothesis and the origin of societal evolution. In Terano, T., Kita, H., Takahashi, S., & Deguchi, H. (Eds.), Agent-Based Approaches in Economic and Social Complex Systems V. Springer Japan, 81–91. 10.1007/978-4-431-87435-5_7 Vromen, J. (2008). Ontological issues in evolutionary economics: The debate between generalized Darwinism and the continuity hypothesis (Working Paper No. 0805). Papers on Economics and Evolution. https://www.econstor.eu/handle/10419/31794 Witt, U. (2006). Evolutionary concepts in economics and biology. Journal of Evolutionary Economics, 16(5), 473–476. 10.1007/s00191-006-0034-4 Witt, U., & Cordes, C. (2007). Selection, learning and Schumpeterian dynamics: A conceptual debate. In Hanusch, H. & Pyka, A. (eds.), Elgar Companion to Neo-Schumpeterian Economics. Edward Elgar Publishing, 316.

146

11 COMPUTATIONAL EVOLUTIONARY ECONOMICS Minimal principle and minimum intelligence Shu-Heng Chen

11.1

Backgrounds and motivation

This chapter is written from the perspective of agent-based modeling (ABM, hereafter); it intends to be a contribution to a subject pertinent enough to connect ABM and evolutionary economics (EE, hereafter). Agent-based computational economics (ACE, hereafter) has often been considered a computational companion to EE or computational evolutionary economics (Pyka and Fagiolo, 2007; Gräbner, 2016), especially after ACE has been buttressed by incorporating the evolutionary computation paradigm (Chen and Chie, 2006; Chie and Chen, 2013). However, if we are not satisfied with just the general notion of change, adaptation, or learning, but prefer to ground ACE upon a deeper level of natural selection,1 further to the extent of the debate between Charles Lyell (1797–1875) and Charles Darwin (1809–1882) on the plausibility of the fish-to-man shift (Dixon and Radick, 2009), then there is one critical yet neglected thread connecting ACE and EE. For the convenience of the subsequent discussion, we shall christen the lessnoticed thread the minimal principle (MinP, hereafter), and, as an illustration, this chapter will focus on one reification of the MinP, namely, minimum intelligence (MI, hereafter) in ACE. This explains the title of the chapter. Before we proceed further, let us first motivate the term, the MinP (Section 11.1.1), and we shall see the pertinence of the MinP to Darwinism from a hierarchical perspective (Section 11.1.2).

11.1.1 Minimum principle The MinP in ACE means that agent-based modelers should minimize the instructions or interventions built into the model and need only to involve the most elementary units or objects which may germinate themselves into high-level, sophisticated, complex objects via the force of nature, namely, Darwinian evolution. Objects set up in this fashion are broadly known as autonomous agents in the ACE community. Hence, by the MinP, modelers will not interfere in the work which is supposed to be taken care of by “nature”; instead, they only invite “nature” to their lay environment (model) and leave the adapted complexity of organisms (agents) in the hand of nature.

DOI: 10.4324/9780429398971-13

147

Shu-Heng Chen

The application of the MinP makes ACE become the computational simulation of Darwinian evolution in the context of economics; it, therefore, connects ACE and EE in their common pursuit for the understanding of increasing adapted complexity or the transition from simple organisms to complex organisms. If we take ACE as a computational EE, which in turn takes Darwinian evolution seriously (Dopfer, 2005; Hodgson and Knudsen, 2010), then the MinP should be given a paramount consideration in ACE.

11.1.2

Hierarchies

The fish-to-man shift, as addressed by Darwin, could be a manifestation of the MinP in biology (Dixon and Radick, 2009). What one needs to know is how simple organisms can evolve into complex forms, such as from prokaryotes to eukaryotes, from single-cellular organisms to multi-cellular organisms, and from the simple labor division of cells to a sophisticated labor division of cells. In this regard, the botanist Boris Mikhaylovich KozoPolyansky (1890–1957) and the evolutionary biologist Lynn Margulis (1938–2011) proposed their theory of symbiogenesis as a supplement to the Darwinian natural selection (Kozo-Polyansky, 2010; Margulis and Fester, 1991). By the theory of symbiogenesis, evolution is not just in the form of selection via competition, but also in the form of variations brought by collaborations. The latter motivates various symbiotic processes, which in turn provide the key mechanism to understand how simple organisms find their way up to complex organisms.2 Herbert Simon (1916–2001) characterized the architecture of complex systems as being hierarchical (Simon, 1962). In this hierarchy, each unit at any given level, together with some neighboring units, becomes the constituents of a unit at the level immediately above and in the meantime is also the composition of many units at the level immediately below; in this recursive manner, the hierarchy is potentially open-ended in both of its ceiling and floor. The complexity of an organism or a system can then be measured by its depth, i.e., the number of levels. Simon further justified hierarchization from an evolutionary viewpoint. Using his invented parable of the competition between two watchmakers, namely, Hora and Tempus, he argued how Hora can take advantage of modularizing the production and, accordingly, encapsulating the completed part. The encapsulated modules can be repeatedly used in later production. This mechanism minimizes the adverse effects resulting from incessant disruptions coming in the middle of production. Simon’s hierarchy hypothesis and the symbiogenesis hypothesis enable us to imagine the possible evolutionary route ungirding the fish-to-man shift. While along this line one can further delve into the literature regarding major transitions in evolution (Jagers op Akkerhuis, 2016), what we need here is how the mechanisms which we learned from these hypotheses can be computationally formulated and operated in ACE. Fortunately, from the 1960s to 1980s, a number of protégés of John von Neumann (1903–1957) in the renown Michigan School pioneered biologically inspired computer science and developed variants of evolutionary computation which can simulate Simon’s hierarchy and mimic some of the related symbiotic-like processes (Burks, 1997; Holland, 1995; Koza, 1992). For example, genetic programming provides us with a transparent layout to see how simple organisms (“atoms”) develop into complex organisms by recursively encapsulating existing modules as part of them (Koza, 1992). In genetic programming, the elementary unit to which Darwinian evolution is applied is modules, or in John Davis’s term (Davis, 2013), the rules, the codes, or the culture genes. Taking rules as the elementary units of evolution 148

Computational evolutionary economics

provides us with flexibility to cope with the evolution of cultures, organizations, and institutions in social sciences.3 These modules initially are simple (primitive), but have the potential to germinate and evolve into complex forms; one, therefore, can possibly simulate the evolution route from simple to complex at an abstract level using genetic programming, which, technically, endorses the MinP. Despite its strong connection with evolution, the minimum principle has not been so well received as opposed to the other two principles in ACE, namely, the simplicity principle (SP, hereafter) and the maximum entropy principle (MEP, hereafter). In fact, the minimum principle may be mistaken as the SP or MEP without noticing their different intellectual origin. In the next section, we briefly review SP and MEP and discuss the relation among the three.

11.2

Simplicity principle

The simplicity principle or the principle of parsimony is frequently taken as a “virtue” of modeling by the tradition of Occam’s razor (also, Ockham’s razor) (Sober, 2015), which was later espoused by statisticians and was extended into various information criteria for the purpose of complexity regularization (Konishi and Kitagawa, 2008). In this enduring practice, the SP basically says that no additional efforts (causes, variables, assumptions, etc.) should be added to the model (explanation) unless it can be economically justified, i.e., if it can be claimed that the benefit (model accuracy) received is greater than the cost (model complexity) incurred. The main challenge of reifying the SP is the setting of the cost function associated with the complexity or the size of the model, even if one could eschew the choice of the statistical loss function.

11.2.1 Minimum description length When equation-based models are replaced with agent-based models and when data-centric models are replaced with thought-centric models, the aforementioned information criteria become even more challenging to apply; however, the essence of the SP remains the same and has been made popularly known as the army slogan “keep it simple, stupid” or the KISS principle (Axelrod, 1997a). The KISS principle does not have a formal procedure, but from its practices in ACE, one can give it a formal notion based on Kolmogorov complexity (Li and Vitányi, 2008). Kolmogorov complexity is proposed to deal with an informal notion of “complexity” in mathematical theory of computation. It measures the amount of information necessary to describe a single object; therefore, it is also termed the minimum description length. For ABM the minimum description length (MDL, hereafter) is very intuitive and practical since an agent-based model itself is an executable computer code and the size (description length) is immediately observable. A code with only one line is intuitively simpler than a code with 1,000 lines; likewise, a behavioral rule with a one-line code is simpler than a behavioral rule described by 1,000 lines. Hence, with the intuitive notion of the MDL (the program-size complexity), the KISS principle becomes not just intuitively clear, but, to a practical extent, measurable. ABMs have two essential components namely, agents, and the space (environment, embeddedness) in which social interactions can happen. The KISS principle can be applied to any of the two components. For example, when applied to the agent component, say, the behavioral rules, the KISS principle directs us to think of rules with a short MDL. A classic example is 149

Shu-Heng Chen

the Tit-for-Tat strategy in the iterated prisoner’s dilemma game (Axelrod, 1997a). The description of Tit-for-Tat is to cooperate on the first move and do whatever the opponent did on the previous move. The code essentially is solely to imitate what the opponent just did, which can be easily coded in one line. In fact, any behavioral rule involving imitation only, such as following the neighbors, the herds, the gurus, the winners, the norms, the conventions, or the routines, is also simple in the sense of MDL, since regardless of what those “leaders” or “constitutions” did, their codes, as a whole or as a part, are taken as a module or subroutine, encapsulated as a symbol, which the followers can easily call them in by one line, such as “do this,” “do that,” etc. This is why imitation is not only prevalent in the world of ABM, but also ubiquitous in the real world; they are viscerally simple.4 A prominent class of ABMs in social science that demonstrate the reification of the KISS principle is the one known as the checkerboard model (Sakoda, 1971; Schelling, 1971; Nowak & May, 1992; Axelrod, 1997b). In these models, agents, situated in simple environment (lattice), follow simple rules to behave. Peter Albin (1934–2008) emblemed this development using the term cellular automata and attributed the intellectual origin to John von Neumann (Albin, 1975; Albin and Foley, 1998). In cellular automata, while agents homogeneously follow simple rules, the system can still generate kaleidoscopically protean patterns, indicating that it does not just operate on the linear summing-up of the simple homogeneity, but incorporates underlying topological information embedded in the space (Wolfram, 2002). This kind of feature has become a tradition in ACE, i.e., to treat the complex economic phenomena being generated by the interactions of simple behaving agents.

11.2.2

Is the SP the MinP?

The two principles, MinP and SP, seem to be identical. All the simple settings in the sense of MDL are, in spirit, consistent with the MinP which dictates the modelers to give only the elementary or primitive settings and leave nature to develop the rest. Furthermore, if the modeler would be interested in the kind of fish-to-man transition as either explananda or explanans, then even for the follower of SP, he/she will respect the force of nature by economically adding to the code a few lines. By this doing, the SP effectively is synonymous with the MinP. In fact, Duffy (2006) justifies the KISS principle by saying, “… the phenomena that emerge from simulation exercises should be the result of multi-agent interactions and adaptation, and not because of complex assumptions about individual behavior … (Ibid, p. 954; italics added).” The rationale given above evinces the consistency of the two principles: the interactions and adaptation of agents are done through nature rather than through complex assumptions of individual behavior. However, here comes Maslow’s hammer: when one is in possession of a hammer, everything starts looking like a nail (Maslow, 1966). By Maslow’s cognitive psychology, the tools originally designed to meet our ends may in turn shape our ends. Maslow warned us that our ends may not be independent of the tools or the principles that we espouse. When one is able to generate and harness complex dynamic patterns or stylized facts using only flat programming, hierarchical programming looks unwieldy. It turns out that, in the practice of SP, the mechanisms related to Simon’s hierarchies, including the formation (encapsulation) of modules (subroutines), imitation by modules, or production of modules by means of modules, are rarely pursued. 150

Computational evolutionary economics

Imitation without the availability of modules makes agents under the SP unable to find simple routes to get complex; hence, in the end, they generally remain to be simple as what they initially are. On the other hand, under the MinP, modules are not supplied to the model as “complex assumptions”; instead, they emerge by dint of the work of nature. The ACE, being computational EE, can, therefore, be much encumbered by the predilection for flat thinking. We shall come back to this point in Section 11.3, but before that let us also look at another popularly mentioned principle in ACE, namely, the maximum entropy principle.

11.2.3

Maximum entropy principle

Entropy has a long history in sciences, originated from thermodynamics and statistical physics by Ludwig Boltzmann (1844–1906) around the 1870s, and has been introduced into theory of communication and information by Claude Shannon (1916–2001) (Shannon and Weaver, 1949). Edwin Jaynes (1922–1998) further formulated it into the maximum entropy principle in probability theory and statistics (Jaynes, 1957). The maximum entropy principle (MEP, hereafter) says that the best distribution for the set of the data of interest is the one that maximizes the entropy subject to a number of constraints revealed by the data. The original formalism of the MEP is rather delicate (Golan, Judge, and Miller, 1997); the main idea is to ask the modeler to minimize his/her degree of prejudice subject to what is objectively known to him/her or to be “maximally non-committal with regard to missing information” (Jaynes, 1957, p. 620). The MEP has also been extensively applied to statistical inference (Golan, Judge, and Miller, 1997). It can be related to the SP; for example, Rissanen (1989) showed that the MEP can be reformulated by the MDL criterion in a more generalized situation, where the number of data constraints is unknown. Like the SP, the MEP has no direct application to ACE. However, what we often see in practice is the application of the entropy-maximizing distribution to the aforementioned two mainstays of ACE. Since the entropy-maximizing distribution in a compact space is a uniform distribution, the uniform distribution is greatly applied to the two mainstays of ACE, especially after they have been properly parameterized. As Jaynes advised, if we know nothing about these agents and their networks, then logically we should reflect our unknowledge through maximum degree of randomness. The MEP is simple in the sense of MDL. However, here is the catch. The randomization procedure itself, say, the random-number generator, has to be already modularized and can be directly used without further efforts; otherwise, depending on the quality of the random numbers generator, the program length may vary (Gentle, 2003), and the MEP may be substantially deviated from the SP. How is the MinP related to the MEP? One essence of the MinP is to leave nature to do the work; however, by Darwinism, a key for nature to do the job is diversity. Since the MEP makes the diversified population more accessible, in this sense, MinP and MEP share a similar pursuit. However, the MEP alone cannot form modules, without this element, we are wagering on a monkey who can type out the complete works of William Shakespeare. It will be hard for agents to get complex with a randomization mechanism only. Without invoking modularity or Simon’s hierarchy; neither the SP nor the MEP will do justice to the MinP. These juxtapositions show that the MinP is unique, paving a unique road to computational EE. 151

Shu-Heng Chen

11.3 Minimum intelligence in ACE Among the two mainstays of ACE, the one considered as the centerpiece is agent engineering. With regards to the design of cognitive agents, the MinP is reified as minimum intelligence (MI). We shall briefly recount the origin of MI in ACE and show that, in addition to the MinP, MI can be also interpreted in terms of the SP and MEP (Section 11.3.1). Furthermore, the idea of MI generally recognized by the ACE community is confined to the “conventional” SP and MEP; however, MI in the sense of the MinP can enrich our understanding of MI and its contribution to Computational EE (Section 11.3.2).

11.3.1

Minimum intelligence by principles

The idea of MI or zero-intelligence (ZI, hereafter) was first introduced to ACE by Gode and Sunder (1993) and has been widely accepted by the ACE community (Ladley, 2012; Iori and Porter, 2018). According to Sunder (2004), the idea popped out in his class on programmed trading in a double-auction environment, where students “pressed us for our own trading strategy so they could trade against it and – beat it. … Toward the end of the term, we finally wrote a trading strategy. … later labeled ‘zero-intelligence’ (ZI) traders … (Ibid, p. 511–512; Italics added).” Since its inception, the MI/ZI agent has been frequently characterized by random behaving, which involves no memory and hence no learning in decision making. Based on the brief description above, MI/ZI can be interpreted as a reification of the SP in the sense of MDL. Here is how Sunder (2004) narrated the ZI strategy. The strategy consists of one line of computer code: if you are a seller with a cost of, say, 40, pick a uniformly distributed random number between 40 and 200 and submit it as an “ask”; if you are a buyer with a value of, say 135, pick a uniformly distributed random number between 0 and 135 and submit it as a “bid.” (Ibid, p. 512; Italics added) This program, the one-line computer code, is even shorter than the background-player strategy, one of the shortest programs submitted to the Santa Fe Double Auction Tournament and also the champion of the tournament (Rust, Miller, and Palmer, 1994). MI/ZI can also be interpreted as a reification of the MEP, since it has another operational meaning, i.e., bid or ask with uniform distribution over a finite interval. In the case of double auction, if the only fact objectively known to us is that no trader can accept a trade with loss, then, subject to this constraint, the entropy-maximizing design is the indeed the MI/ZI strategy described above. Here, the SP can coincide with the MEP when the entropy-maximizing behavior can be programmed into the “one-line code” as in the case of Gode and Sunder.5 This coincidence makes the MEP, sometimes, loses its independence; the maximum entropy agent is often termed as the zero-intelligence agent when it can be better interpreted as the MEP agent in Jaynes’s sense. For example, Wright (2009) stated “[s]o rather than taking an ‘explicit microfoundations’ approach, in which individuals are represented as ‘white-box’ sources of fully-specified optimizing behavior (‘rational agents’), we instead represent individuals as ‘black box’ sources of unpredictable noise subject to objective constraints (‘zero-intelligence agents’)” (Ibid, Abstract; Italics added). Finally, MI/ZI can also be interpreted as a reification of the MinP. From the viewpoint of psychology, MI/ZI denotes the minimum degree of deliberation or the minimum cognitive

152

Computational evolutionary economics

effort. It is so minimal that one can even cross the borderline between humans and animals to evoke the early animal experiments of cats (Thorndike, 1911) or chimpanzees (Köhler, 1925). In these experiments, analogous to randomly behaving, animals acted either by random trails, e.g., Thorndike’s cats, or by brutal force, e.g., Köhler’s chimpanzees. For the latter, some degree of deliberation was observed only after the constant failures of using brutal forces. Therefore, if the device of MI/ZI is interpreted as the initial inputs, elementary enough to leave nature to have a role to play, then MI/ZI can be taken as an exemplar of the MinP. In this exemplar, the ACE modeler is doing a computational counterpart of animal experiments. The work of nature, in the computational sense, refers to the interaction between agents (cats, chimpanzees) and their embeddedness (the cage and the bananas outside the cage). What emerges from these interactions is the observed intelligent behavior (circumventing the cage), which we modelers do not impose as complex assumptions in the first place. We allow nature to present the emerged intelligence to us as those psychologists were yearning to see. By the same analogy, we, as a modeler, are also curious to know how the emerged intelligence was formed (how the cats or the chimpanzees found their way out). Nevertheless, unlike psychological experiments, we modelers can archive everything about the agents during the simulation, including their “stream of consciousness” (Chen and Venkatachalam, 2017), and, from there, we can see whether our agents (cats, chimpanzees) found something unknown to us (Chen and Yu, 2011). In sum, we see that MI/ZI can be simultaneously interpreted as the reification of all three principles, the SP, MEP, and MinP. In the next section (Section 11.3.2), we shall extend the juxtaposing analysis made in Section 11.2 to the case of MI and examine what MI could mean for shaping the development of ACE and its relation to Computational EE.

11.3.2

Minimum intelligence in practice

While MI can be interpreted by three principles, in practice the interpretation is mainly slanted toward the SP and the MEP, especially when the two are coinciding with each other as noted above. To see this, it is useful to have a broader review of the design of cognitive agents in ACE. The term “intelligence” in ACE was first observed in models of learning behavior when Homo Economicus is replaced with Homo Sapiens. This stream of studies was further inspired by experimental economics (Arthur, 1993; Arifovic, 1994; Rust, Miller, and Palmer, 1994; Roth and Erev, 1995; Axelrod, 1997a). Some of the experiments recruited “avatars” (human-written programs), by which human behavior is directly observable and analyzable. Gigerenzer and Selten (2001) and Sargent (1993) offered toolboxes that can be used to construct these avatars; Duffy (2006) surveyed these tools and arranged them by their complexity in the sense of the MDL; from the low end to the high end, there are the zero-intelligence (near-zero-intelligence, zero-intelligence plus), generalized reinforcement learning, neural nets, and evolutionary algorithms. The design of the cognitive agent can be viewed as a choice from the indicated toolbox, and this choice is often influenced by the KISS principle and the cellular automaton tradition (Section 11.2.1). Gode and Sunder (1993) carried on this tradition, as they convincingly stated: “Adam Smith’s invisible hand may be more powerful than some may have thought: when embodied in market mechanisms such as a double auction, it may generate aggregate rationality not only from individual rationality but also from individual irrationality” (Gode and Sunder, 1993, p. 136; Italics added). This lesson reiterates Adam Smith’s invisible hand and Friedrich Hayek’s cosmos: order is self-organized, emerging, and unintended by individuals. 153

Shu-Heng Chen

In the vein of the cellular automaton tradition, one finds many reifications of SP/MEP in the literature of ACE, in particular, in macroeconomics and financial markets. For example, Wright (2005, 2009) built an agent-based circular-flow model. All agents, households, and firms are of the Gode-Sunder type, uniformly randomly making their decisions in both the goods market and the labor market; the interactions of these MI/ZI agents can, nonetheless, generate interesting features related to business cycles, income/wealth distribution, firm-size distribution, etc. Another prominent exemplar is manifested in the response of the ACE community to what Kenneth Boulding (1910–1993) had evinced more than six decades ago. In the larger processes of recession and recovery, inflation and deflation, we have already noticed how an image on the part of one person of shall we say an impending depression produces behavior which has the effect of making that depression all the more probable. These phenomena are part of a wider class of processes which might be called chainreaction processes. Fashion, whether in clothes or ideas, is an interesting example of process. Many of these processes are very imperfectly understood, … Some light is thrown on this problem by work which has been done on models of the spread of epidemics. (Boulding, 1956, p. 124; Italics added) In the quotation above, Boulding conceived of “social epidemics” as the underpinnings of macroeconomic activity; he pointed out the mathematical epidemics, such as the Kermack-McKendric Model established then, can shed light on the formation and the spread of ideas, opinions, sentiments, etc. and its impacts on the entire economy. He even attempted to establish a new science, called Eiconics.6 At the turn of this century, two new fields were burgeoning; they are socioeconomics (Prechter, 1999) and social economics (Durlauf and Young, 2001). These new fields inherit part of Eiconics, albeit with a much more concentrated scope. Recently, in Robert Shiller’s narrative economics (Shiller, 2019), not only do we see the applications of the Kermack-McKendric model to social epidemics, but also those of the ABM. While the ABM of social epidemics is still developing, due to its open-ended nature, it has the potential to reach the chain-reaction processes dubbed by Boulding. One of its promising applications is in agent-based financial markets, which relates the stylized facts observed in the financial markets to the underlying opinion or sentiment dynamics (Chen, Chang, and Du, 2012). Agents in these models are normally assumed to be MI, having a finite number of opinions (optimism vs. pessimism, fundamentalists vs. chartists) and/or a finite number of investment strategies (active or passive); the numbers are usually small and the options are given. This setting simplifies the decision problem as a typical stochastic discrete choice. Being MI, agents make decisions stochastically, but the probability for each choice is distorted by agents’ experience, characterized by variants of reinforcement learning (Chen, 2013). Reinforcement learning originates in animal experiments (Thorndike, 1911), which are both psychologically and algorithmically simple. With this slight enhancement, ZI can be upgraded to its various plus versions, known as ZI-Plus (ZIP), which flag the cellular automaton tradition. It can be shown that the interactions of these ZIP agents are able to generate many interesting features of financial markets, such as excess volatility, fat-tailed distribution of returns, volatility clustering, etc. (Chen, Chang, and Du, 2012). Under the cellular automaton tradition, other high-end tools are rarely considered when order, patterns, or structure can already be grown by simple or random agents, but this 154

Computational evolutionary economics

success is at the cost of leaving nature substantially idle and all simple entities simple throughout the simulation. Learning, in a nutshell, is just about regime switching or number crunching, nothing accumulated as culture: no modules, no hierarchies, no transformation, and no role for the MinP. There is an occasion when we see the necessity of invoking the high-end tools, which is due to the surging interest in the dialog and cooperation between ACE and psychology, neuroscience, and cognitive science (Epstein, 2014; Chen, 2016). In this scenario, we are concerned with the effect of intelligence as an attribute of agents to the aggregate performance of a system (market or society) in which these agents participate, and their wellbeing, such as earning capacity, wealth share, etc. A typical question is whether agents characterized by low-end tools will end up with a relatively inferior position when compete with agents using high-end tools, or, on the contrary, whether smart agents will be driven out by MI/ZI agents (Chen, Zeng, and Yu, 2009; Chen and Tai, 2010; Tai, Chen, and Yang, 2018). In this scenario, the low-end tools are compared with high-end tools like genetic programming. With the presence of the former, the involvement of the latter is hardly considered as MI, and certainly not the reification of the SP or the MEP. Nevertheless, as we shall see in the next section, these high-end tools, neural nets or genetic programming, can be perceived as MI; they are, instead, the reification of the MinP. We shall focus on neural nets, but the general lesson applies to genetic programming, namely, the negligence of modularity inevitably prevents us from seeing the force of nature and limits our understanding of MI.

11.4

Neural nets and minimality

Are neural nets also a kind of MI, if so, how? Talking about MI, one notices that the debate between symbolism and connectionism in the history of cognitive science may be recognized as a contrast between the top-down vs. bottom-up approach to cognition and intelligence (Minsky and Papert, 1988). Framed by these contrasts, if one chooses to view intelligence as an emergent property of a complex system, say, the brain, then connectionism formulated in ABM actually is a reification of the MinP by getting on with MI. Moreover, its advent symbolizes the dawn of the cellular automaton tradition. From the cognitive-science viewpoint, the cellular automaton tradition could be further traced back to McCulloch-Pitts neural nets (McCulloch and Pitts, 1943; Albin, 1975). The McCulloch-Pitts neural net (MPNN, hereafter) was originally proposed as a computational model of brain which can basically do all propositional or predicate calculus as demonstrated by Principal Mathematica (Whitehead and Russell, 1910, 1912, 1913). While the relation between the McCulloch-Pitts neural nets and ABM is not immediately clear, the subsequent development in the direction of Boolean networks or a society of finite state automata by Stuart Kauffman (Kauffman, 1993) clearly shows that MPNNs can be accepted as a kind of cellular automata. Von Neumann had reviewed MPNNs in contrast to the Turing machines. He observed that McCulloch and Pitts described structures which are built up from very simple elements, so that all you have to define axiomatically are the elements, and then their combination can be extremely complex. Turing started by axiomatically describing what the whole automaton is supposed to be, without telling what its elements are, just by describing how it’s supposed to function. (Von Neumann, 1966, p. 43; Italics added) 155

Shu-Heng Chen

Von Neumann’s own machine, the 29-state automata, was also under the influence of the MPNN. Since the inception of the MPNN, neurons as the elementary units have remained to be simple or minimal intelligent as the MinP evinces, and nature will take over from here. Our focus should be riveted on the formation of underlying connections or organizations; these topologies are often not pursued by the SP and the MEP, but they transfer MI into high-end intelligence, alluding to the kind of fish-to-man transition. Donald Hebb (1904–1985) took the assembly of neurons (modules), instead of neurons, as a fundamental building block, upon which assemblies of assemblies (networks of networks) can develop with his Hebb learning rule (Hebb, 1949). Hebb’s learning indicates how the connections among agents (neurons) may dynamically change or evolve so that the assembly (network) or Simon’s hierarchies can be formed bottom-up. As a behavior rule, Hebb learning involves only simple reinforcement mechanism as a habitual process, comparable to the homophily principle (Axelrod, 1997b): agents only interact with those alike and these interactions will make the alike more alike. The process beginning with elementary simple units and leaving the nature to selforganize them into modules also guided Frank Rosenblatt (1928–1971) in his work on perceptron (Rosenblatt, 1962). It is significant that the individual elements … have never been demonstrated to possess any specifically psychological functions, such as … “intelligence.” Such properties, therefore, presumably reside in the organization and functioning of the network as a whole … In order to understand how the brain works, it thus becomes necessary to investigate the consequences of combining simple neural elements in topological organizations analogous to that of the brain. (Ibid, 1962, pp. 9–10; Italics added) As compared to the MPNNs, what Rosenblatt proposed is a learning rule that can help machine to learn how to do propositional calculus, which, was, instead, designed manually in the MPNNs. Rosenblatt himself differentiated his idea from McCulloch and Pitts’s by calling the latter monotypic models and his own genotypic models (Ibid, Sections 11.2.2 and 11.2.3). In terms of agent-based modeling, agents in the MPNN follow fixed given rules, but agents in Rosenblatt’s perceptron are more autonomous by being endowed with a rule to learn from feedback. Neither is significantly different from other MI/ZI used in ACE; the former is similar to Wolfram’s elementary cellular automata (Wolfram, 2002), and the latter is similar to Axelrod’s model of cultural transmission (Axelrod, 1997b).

11.5 Concluding remarks In this chapter, we review the three principles that are applied to work with ABM/ACE. Among the three, the simplicity principle and the maximum entropy principle already existed in the time of equation-based modeling. However, ABM differs from equation-based modeling in many aspects. It, therefore, raises the question of whether a different principle is needed, and how this principle differs from the existing ones, specifically, its implications for economics. In this chapter, the minimal principle is proposed as the third possibility. We argue throughout the chapter that when ACE is considered computational evolutionary economics, the molders should minimize their inputs to the model so as to induce the rest as 156

Computational evolutionary economics

much as possible from the force of nature (Darwinian evolution). This is the way that Darwinism in economics can be computationally reified. From this viewpoint, the SP and the MinP share the same pursuit for minimum inputs, but the inputs (the agents, the objects) of the former are expected to be simple throughout the simulation, whereas those of the latter are much unleashed. This distinction sets the MinP apart from the SP as well as the MEP. Mechanisms related to assemblies, encapsulations, modulus, and hierarchies are not further pursued by the SP in practice, nor by the MEP in its reifications. Even though imitation is permissible by the SP and, to some extent, by the MEP, practitioners of the SP and MEP normally will not use this mechanism to recursively build hierarchies, produce modules by means of modules, or create networks and culture. The MinP, on the other hand, constantly uses imitation, reproduction, copying, and mutation, those simple mechanisms, to form modules, their assemblies, networks, and organizations, as reviewed in Section 11.4. Hence, simplicity is not an absolute idea, not independent of the availability of modules, nor is it independent of social learning, networking, and culture, which altogether in the end simplify our otherwise complex life (Henrich, 2016). The absence of the mechanism for modules and hierarchies can limit the exposure of EE to ACE. For example, the signs in Peirce’s semiotics are distinguished, in different stages of evolution, by icons, indices, and symbols. When they are applied to Boulding’s social epidemics or Shiller’s narrative economics, we are addressing “viruses” that are layered, which is more general than just switching among finite discrete choices. We can certainly stick to the SP, and “simplify” the social epidemics or opinion dynamics into the switch among a finite number of choices. This kind of simplicity has its virtue as we recount in Section 11.3.2, but when we face Veblen’s opening message of Chapter VIII of Theory of the Leisure Class (Veblen, 1899/2007), it seems to be rather constrained. The life of man in society, just like the life of other species, is a struggle for existence, and therefore it is a process of selective adaptation. The evolution of social structure has been a process of natural selection of institutions. The progress which has been and is being made in human institutions and in human character may be set down, broadly, to a natural selection of the fittest habits of thought and to a process of enforced adaptation of individuals to an environment which has progressively changed with the growth of the community and with the changing institutions under which men have lived. (Ibid, p. 125; Italics added)

Acknowledgments Research grants Taiwan MOST 106-2420-H-004 -011-MY2 and MOST 108-2410-H-004 016 -MY2 are gratefully acknowledged.

Notes 1 In this chapter, when the term Darwinism, Darwinian evolution, or natural selection is mentioned, it is referred to what has been suggested as Neo-Darwinism ( Cziko, 1997; Mayr, 1988), universal Darwinism, or generalized Darwinism ( Hodgson and Knudsen, 2010) covering various genetic mechanisms and multi-level co-evolution. 2 In evolutionary biology, this is known as the hierarchical theory ( Eldredge et al., 2016).

157

Shu-Heng Chen 3 See also Dopfer (2004). 4 Of course, here, we assume that the choice of target to follow and the act to imitate is straightforward; otherwise, one needs to lengthen the code to describe how the choice is made and how the imitation is carried out, which may make imitation be grotesquely complex. 5 As we remark in Section 11.2.3, this coincidence rests upon the assumption that the related randomization procedures are available as modules and can be directly included. 6 The significance of icons in evolutionary theory, in fact, has already been expounded by Charles Peirce (1839–1914) in his established semiotics.

References Albin, P. S. (1975). The analysis of complex socioeconomic systems. Lexington, MA: Lexington Books. Albin, P. S., & Foley, D. K. (1998). Barriers and bounds to rationality: Essays on economic complexity and dynamics in interactive systems. Princeton, NJ: Princeton University Press. Arifovic, J. (1994). Genetic algorithm learning and the cobweb model. Journal of Economic Dynamics and Control, 18(1), 3–28. Arthur, W. B. (1993). On designing economic agents that behave like human agents. Journal of Evolutionary Economics, 3(1), 1–22. Axelrod, R. (1997a). The complexity of cooperation. Princeton University Press. Axelrod, R. (1997b). The dissemination of culture: A model with local convergence and global polarization. Journal of Conflict Resolution, 41(2), 203–226. Boulding, K. E. (1956). The image: Knowledge in life and society. University of Michigan Press. Burks, A. W. (1997). Logic, learning, and creativity in evolution. In Houser, N., Roberts, D. D., & Van Evra, J. (Eds.). Studies in the logic of Charles Sanders Peirce (pp. 497–534). Indiana University Press. Chen, S. H. (2013). Reasoning-based artificial agents in agent-based computational economics. In Nakamatsu, K., & Jain, L. (Eds.). The handbook on reasoning-based intelligent systems (pp. 575–602). Chen, S. H. (2016). Agent-based computational economics: How the idea originated and where it is going. Routledge. Chen, S. H., & Chie, B. T. (2006). A functional modularity approach to agent-based modeling of the evolution of technology. In Namatame, A., Kaizouji, T., & Aruka, Y. (Eds.). The complex networks of economic interactions (pp. 165–178). Berlin, Heidelberg: Springer. Chen, S. H., & Tai, C. C. (2010). The agent-based double auction markets: 15 years on. In Takadama, K., Cioffi-Revilla, C., & Deffuant, G. (Eds). Simulating interacting agents and social phenomena (pp. 119–136). Tokyo: Springer. Chen, S. H., & Venkatachalam, R. (2017). Agent-based modelling as a foundation for big data. Journal of Economic Methodology, 24(4), 362–383. Chen, S. H., & Yu, T. (2011). Agents learned, but do we? Knowledge discovery using the agent-based double auction markets. Frontiers of Electrical and Electronic Engineering in China, 6(1), 159–170. Chen, S. H., Chang, C. L., & Du, Y. R. (2012). Agent-based economic models and econometrics. The Knowledge Engineering Review, 27(2), 187–219. Chen, S. H., Zeng, R. J., & Yu, T. (2009). Co-evolving trading strategies to analyze bounded rationality in double auction markets. In Riolo, R., Soule, T., & Worzel, B. (Eds.). Genetic programming theory and practice VI (pp. 1–19). Boston, MA: Springer. Chie, B. T., & Chen, S. H. (2013). Non-price competition in a modular economy. An agent-based computational model. Economia Politica, 30(3), 273–300. Cziko, G. (1997). Without miracles: Universal selection theory and the second Darwinian revolution. MIT Press. Davis, J. B. (2013). The emergence of agent-based modeling in economics: Individuals come to bits. Filosofia de la Economia, 1(2). Dixon, M., & Radick, G. (2009). Darwin in Ilkley. The History Press. Dopfer, K. (2004). The economic agent as rule maker and rule user: Homo Sapiens Oeconomicus. Journal of Evolutionary Economics, 14(2), 177–195. Dopfer, K. (Ed.). (2005). The evolutionary foundations of economics. Cambridge University Press.

158

Computational evolutionary economics Duffy, J. (2006). Agent-based models and human subject experiments. In Tesfatsion, L., & Judd, K. (Eds), Handbook of computational economics: Agent-based computational economics (Vol. 2, pp. 949–1011). Oxford: Elsevier. Durlauf, S. N., & Young, H. P. (Eds.). (2001). Social dynamics. MIT Press. Eldredge, N., Pievani, T., Serrelli, E., & Tëmkin, I. (Eds.). (2016). Evolutionary theory: A hierarchical perspective. University of Chicago Press. Epstein, J. M. (2014). Agent_Zero: Toward neurocognitive foundations for generative social science. Princeton University Press. Gentle, J. E. (2003). Random number generation and Monte Carlo methods (Vol. 381). New York, NY: Springer. Gigerenzer, G., & Selten, R. (Eds.). (2001). Bounded rationality: The adaptive toolbox. MIT Press. Gode, D. K., & Sunder, S. (1993). Allocative efficiency of markets with zero-intelligence traders: Market as a partial substitute for individual rationality. Journal of Political Economy, 101(1), 119–137. Golan, A., Judge, G., & Miller, D. (1997). Maximum entropy econometrics: Robust estimation with limited data. John Wiley and Sons. Gräbner, C. (2016). Agent-based computational models–a formal heuristic for institutionalist pattern modelling? Journal of Institutional Economics, 12(1), 241–261. Hebb, D. O. (1949). The organization of behavior: A neuropsychological theory. Psychology Press. Henrich, J. (2016). The secret of our success: How culture Is driving human evolution, domesticating our species, and making us smarter. Princeton University Press. Hodgson, G. M., & Knudsen, T. (2010). Darwin’s conjecture: The search for general principles of social and economic evolution. University of Chicago Press. Holland, J. H. (1995). Hidden order: How adaptation builds complexity. Reading, MA: AddisonWesley. Iori, G., & Porter, J. (2018). Agent-based modelling for financial markets. In Chen, S. H., Kaboudan, M., & Du, Y. R. (Eds.). The Oxford handbook of computational economics and finance. Oxford University Press. Jagers op Akkerhuis, G. A. J. (Ed.). (2016). Evolution and transitions in complexity: The science of hierarchical organization in nature. Springer. Jaynes, E. T. (1957). Information theory and statistical mechanics. Physical Review, 106(4), 620–630. Kauffman, S. A. (1993). The origins of order: Self-organization and selection in evolution. Oxford University Press. Konishi, S., & Kitagawa, G. (2008). Information criteria and statistical modeling. Springer Science & Business Media. Koza, J. R. (1992). Genetic programming: On the programming of computers by means of natural selection. MIT Press. Kozo-Polyansky, B. M. (2010). Symbiogenesis: A new principle of evolution. Harvard University Press. Russian origin, 1924. Köhler, W. (1925). The mentality of apes. London: Kegan. Ladley, D. (2012). Zero intelligence in economics and finance. The Knowledge Engineering Review, 27(2), 273–286. Li, M., & Vitányi, P. (2008). An introduction to Kolmogorov complexity and its applications. New York, NY: Springer. Margulis, L., & Fester, R. (Eds.). (1991). Symbiosis as a source of evolutionary innovation: Speciation and morphogenesis. MIT Press. Maslow, A. H. (1996). The psychology of science: A reconnaissance. Harper & Row. Mayr, E. (1988). Toward a new philosophy of biology: Observations of an evolutionist. Harvard University Press. McCulloch, W. S., & Pitts, W. (1943). A logical calculus of the ideas immanent in nervous activity. The Bulletin of Mathematical Biophysics, 5(4), 115–133. Minsky, M., & Papert, S. A. (1988). Perceptrons: An introduction to computational geometry. MIT Press. Nowak, M. A., & May, R. M. (1992). Evolutionary games and spatial chaos. Nature, 359(6398), 826–829. Prechter, R. R. (1999). The wave principle of human social behavior and the new science of socioeconomics. New Classics Library.

159

Shu-Heng Chen Pyka, A., & Fagiolo, G. (2007). Agent-based modelling: A methodology for neo-Schumpeterian economics. In Hanusch, H., & Pyka, A. (Eds.). Elgar companion to neo-Schumpeterian economics (pp. 467–487). Edward Elgar. Rissanen, J. (1989). Stochastic complexity in statistical inquiry. World Scientific. Rosenblatt, F. (1962). Principles of neurodynamics: Perceptrons and the theory of brain mechanisms. Spartan Books. Roth, A. E., & Erev, I. (1995). Learning in extensive-form games: Experimental data and simple dynamic models in the intermediate term. Games and Economic Behavior, 8(1), 164–212. Rust, J., Miller, J. H., & Palmer, R. (1994). Characterizing effective trading strategies: Insights from a computerized double auction tournament. Journal of Economic Dynamics and Control, 18(1), 61–96. Sakoda, J. (1971). The checkerboard model of social interaction. Journal of Mathematical Sociology, 1, 119–132. Sargent, T. (1993). Bounded rationality in macroeconomics. Oxford University Press. Schelling, T. C. (1971). Dynamic models of segregation. Journal of Mathematical Sociology, 1(2), 143–186. Shannon, C. E., and Weaver, W. (1949). The mathematical theory of communication. University of Illinois Press. Shiller, R. (2019). Narrative economics: How stories go viral & drive major economic events. Princeton University Press. Simon, H. (1962). The architecture of complexity. Proceedings of the American Philosophical Society, 106(6), 467–482. Sober, E. (2015). Ockham’s razors: A user’s manual. Cambridge University Press. Sunder, S. (2004). Markets as artifacts: Aggregate efficiency from zero-intelligence traders. In Augier, M., & March, J. G. (Eds.) Models of a man: Essays in memory of Herbert A. Simon. The MIT Press. Tai, C. C., Chen, S. H., & Yang, L. X. (2018). Cognitive ability and earnings performance: Evidence from double auction market experiments. Journal of Economic Dynamics and Control, 91, 409–440. Thorndike, E. L. (1911). Animal intelligence: Experimental studies. Hafner Publishing. Veblen, T. (2007) [1899]. The theory of the leisure class: An economic study of institutions. Edited by Banta, M. Oxford University Press. Von Neumann, J., & completed by Burks, A. (1966). Theory of self reproducing automata. University of Illinois Press. Whitehead, A., & Russell, B. (1910, 1912, 1913). Principia mathematica. Cambridge University Press. Wolfram, S. (2002). A new kind of science. Wolfram Media Inc. Wright, I. (2005). The social architecture of capitalism. Physica A: Statistical Mechanics and its Applications, 346(3–4), 589–620. Wright, I. (2009). Implicit microfoundations for macroeconomics. Economics, 3(1). https://www.degruyter. com/document/doi/10.5018/economics-ejournal.ja.2009-19/html

160

12 EVOLUTIONARY MODELING AND THE RULE-BASED APPROACH Thomas Grebel

12.1 What else if not optimization? It was a major achievement when neoclassical economics was able to offer a complete and consistent methodology, at the beginning of the 20th century. William S. Jevons, Vilfredo Pareto, Léon Walras, and many others contributed to an integrative taxonomy about how to model economic behavior as optimal behavior (Grebel, 2004; Barreto, 1989). The homo oeconomicus, the profit- and utility-optimizing economic actor was born – and at the same time, the critique of the concept. The assumption that economic actors behave in a perfectly rational manner and are able to optimize their economic actions, was and still is a strong simplification of reality. In order to optimize, actors must have perfect information, knowledge about their objective func­ tion, and about all relevant constraints. Otherwise, the state of optimality and its equilibria remain illusive. The critique of the main assumptions and its methodological approach is as old as the neoclassical paradigm. As von Mises (1959, p. 66) argued: “Economics does not follow the procedure of logic and mathematics. It does not present an integrated system of pure aprioristic ratiocination severed from any reference to reality”. Though the optimization hypothesis might be instructive to better comprehend reality, it remains a “statical” rather than a “dynamical” approach (Marshall, 1948, p. 19), because in equilibrium there is no reason for economic actors to change their behavior. Hence, the system comes to a halt and does not correspond to a real, ever-evolving economy (Schumpeter, 1934). In search of an alternative methodological approach to incorporate the dynamics of economic change, many economists pointed toward an evolutionary metaphor. Veblen (1898) asked, “Why is Economics not an Evolutionary Science?” Marshall (1948, p. 19) stated that “The Mecca of the economist [should lie] (…) in economic biology”. Various other economists suggested to use an evolutionary metaphor such as Alfred Marshall, Carl Menger, Friedrich Hayek (Hodgson, 1998), or Schumpeter (1934, 1942) with a particular focus on economic change (Shionoya, 1998). Meanwhile, various heterodox approaches were developed that address the dynamic perspective of non-optimizing economic actors and use an evolutionary metaphor such as

DOI: 10.4324/9780429398971-14

161

Thomas Grebel

“Generalized Darwinism” (Hodgson, 2013), “Organizational Ecology” (Dollimore, 2014), or the “General Theory of Economic Evolution” by Dopfer and Potts (2007) which ex­ plicitly offer a taxonomy of economic change. In this contribution, the focus is put on the rule-based approach by Dopfer and Potts (henceforth DP) who develop an approach that can help simplify evolutionary modeling and is able to reconcile orthodox and heterodox thinking.

12.2 Robinson Crusoe as a rule-user and -maker In 1719, when Daniel Defoe wrote The Life and Adventures of Robinson Crusoe, he could not have been aware of his main character becoming one of the most prominent figures in economics. Many economists, starting already in the mid-19th century, exploited the idea of a solitary man on a deserted island. But in contrast to Dafoe, who invested a lot of time in a detailed description of Robinson’s personality, economists turned him into an excessively simplified figure: the economic man, a pure optimizer on economic grounds. Jevons, Pareto, Edgeworth, and many more of the marginal school used this stereotypical figure to illustrate how Robinson would maximize his subjective utility or minimize costs as a perfectly rational consumer or producer who is faced with scarcity: the homo oeco­ nomicus (Grapard, 1995). In both, the novel and the economists’ model, as Grapard (1995) pointed out, personal characteristics and social ties are neglected. In economic models, Robinson does not struggle with the possibility to end up with suboptimal decisions. He is equipped with perfect capabilities, has complete information, there are no historical events that could influence his decision making, and always achieves his maximum benefit or minimum cost under the given circumstances. Evidently, economic models must be simplified representations of reality. In Robinson’s world, alone on an island, without economic change, no technological progress, no social interactions, and no (social) institutions, utility-/profit-maximizing behavior appears fea­ sible. It seems reasonable to just look at the observed artifacts of Robinson’s economic operations such as consumption or production, instead of asking what actually caused Robinson to decide the way he does. The rule-based approach of DP conceives economic behavior as a result of the cognitive process of a homo sapiens oeconomicus, that is, an economic actor who may also make fallible decisions in contrast to the homo oeconomicus. The type of rules such kind of actor uses, DP categorize according to the classification system of rules as depicted in Table 12.2, which motivates the concept of rules by means of a simplified example given in the following sub-sections.

12.2.1 Robinson in isolation Revisiting Robinson Crusoe’s world, the narrative under the lens of the rule-based approach looks slightly different. For simplicity, let us also neglect any social, cultural, or regulatory norms on Robinson’s island. Then, his only concern was to survive. In the terminology of DP, Robinson would apply the generic rule “try to survive”. The rule is called “generic”, because it suggests no immediate behavioral action on how he could manage to survive. Also, the second rule he might use, i.e. “search and find something to eat” is generic, because it does not provide an operational search strategy, either. Having 162

Evolutionary modeling and the rule-based approach

searched the island in a random manner, Robinson may finally find something to eat, let us assume coconuts. By picking and eating the coconut, Robinson successfully actualizes his main generic rule “try to survive”. In DP’s terminology, Robinson performs an operation on the operant level in the moment he consumes – the level at which actions take an eco­ nomic impact on available resources. In neoclassical terms, the experimental conduct of searching and walking around in a trial-and-error fashion until finding something to eat simply reflects transaction cost. If the economist’s objective is to describe how Robinson copes with scarce resources in solitude, an in-depth analysis of those costs seems negligible. Even the learning process during which Robinson might come up with further new generic rules to better structure his search efforts seem of little interest. Once Robinson has learned the shortest path from his hut to the coconuts, he henceforth manages to minimize transaction costs and will always follow directly the shortest path to the coconuts. In a more rigorous neoclassical manner: Robinson maximizes his utility subject to constraints and ends up in an equilibrium; day in and day out he walks to the coconut place, picks one, eats it, and goes back home. This example describes an economic agent as a rule-user in isolation without social interaction and technical change. Hence, Robinson exclusively employs his own rules, or as DP labels it, Robinson employs subject rules, which DP divide into cognitive (“try to sur­ vive”) and behavioral (“follow the path to the coconuts”) rules.

12.2.2

Institutions and social object rules

When Friday joins Robinson on the island, Robinson has to adapt his behavior. Unless coconuts are abundant, he will face competition for the supply of coconuts, since both are forced to apply the subject-cognitive rule “try to survive”. As soon as Friday learned about the existence of the coconut place and both realized that coconuts are scarce, they cannot help creating a object-social rule to coordinate cohabitation on the island. The options for social object rules are manifold such as “fight for coconuts against the competitor” or “steal from the competitor” or “leave the island and find happiness else­ where”. Suppose both agree to share available coconuts and economize on consumption, they could live together in peace on the island (state of equilibrium). Suppose instead that Robinson claimed ownership of all windfall coconuts, Friday would have to react to this new social object rule. He could either accept (adopt) Robinson’s generic rule, draw the consequences, and create new generic rules, or he could simply ignore it. This illustrates the distinction between the “generic” and the “operant” level in DP’s taxonomy. As long as Friday does not adopt this new rule and keeps on picking windfall coconuts instead, the generic rule has no impact on the operant level. In case Friday agrees, he adopts the new rule and becomes a carrier of that rule. He no longer picks windfall coconuts, but tries to adjust his behavior in order to avoid conflicts with Robinson.

12.2.3

Object rules: Technical change

With an increasing number of economic actors in an economy, economic behavior becomes more complex. Institutional or social object rules, as DP label them, are then required to coordinate social interaction facilitating the coexistence of multiple heterogeneous actors. To give a more comprehensive picture in this respect, Table 12.2 in the appendix illustrates DP’s class of rules. In a world with only two actors, such kind of rules are rather irrelevant. 163

Thomas Grebel

For the purpose of this contribution, we will only focus on technical-object rules. Given that Friday accepts Robinson’s rule, he still needs to pursue the subject rule “try to survive”, necessarily though with a different actualization. One of many conceivable generic rules could be: “try to reach the coconuts on the tree”. This generic rule, in turn, can also lead to additional generic rules which eventually may impact Friday’s economic operations. Friday could invent a technology in order to harvest coconuts on a palm tree. Again, generic rules to solve this problem are abundant. When reflecting on the subject-behavioral rule: “Find something to reach coconuts on the tree”, he could deduce the operational rule “climb the tree”; he could invent a coconut picker whose artifact is a rod with a knife at the end, under which is a basket. This would mean to combine several generic technical-object rules: “1) use a long object in order to reach something far up the tree; 2) use a knife to cut off the coconuts from the tree; and 3) use something to prevent coconuts from reaching the ground and thus becoming windfall fruits”. Note that all generic rules can actualize into different artifacts, because generic rules are bimodal: the knife, for instance, could be made of wood, shell, or stone, the basket could be made out of grass, suitable plants, or the fabric from Robinson’s clothes. With this artifact, i.e. picking rod, Friday can now perform economic operations. He harvests coconuts up in the trees by means of a picking rod. In orthodox terminology, Friday applies a given technology to “produce” coconuts, which he then consumes to maximize his subjective utility.

12.3

Economic evolution

As to economic evolution, the invention of new generic rules and the recombination of generic rules is key. Yet, it remains unclear, whether the invention of a new generic rule or a new combination of existing generic rules impacts economic evolution. If Robinson were the only one to create and adopt the object-social rule that windfall coconuts are his with Friday not following (adopting) that rule, Robinson’s generic rule would have no impact on the overall economic performance – ruling out the option that Robinson could enforce his rule against Friday, vice versa. Hence, the possible adoption of rules by economic actors is decisive in economic evolution. According to DP, the group of actors that adopted a certain generic rule belong to a meso-unit and simultaneously form a meso population. Because every economic actor carries multiple generic rules, everyone belongs to multiple meso populations evolving individually over time. Thereby every generic rule undergoes a meso trajectory: rule origination, ruleadoption, and rule-retention. From a macro perspective, meso units co-exist in an ecosystem, in which meso units are steadily exposed to the process of de-coordination, re-coordination, and stabilization. Some generic rules are short-lived, others become a persistent member of 0th-order constitutive rules. In the case of Robinson and Friday, the mutual acceptance of private property would be an example of such Zeroth-order constitutive rule. Compare Table 12.1 in the appendix. Rules that guide economic operations, i.e. operations on the operant level, belong to the set of first-order operational rules. These rules, when applied by carriers, leave a trail of economic artifacts such as actualized production, consumption, or demand and supply, namely economic artifacts that are of primary interest in mainstream neoclassical eco­ nomics. On the island of Robinson and Friday, the rules such as “follow the shortest path from Robinson’s hut to the palm tree” and “use the invented coconut-picking rod to harvest coconuts”, would reflect transaction or production cost, respectively, in neoclassical 164

Evolutionary modeling and the rule-based approach

economics. As these rules are applied directly to transform goods or resources into value for Robinson and Friday, they are considered first-order operational rules. The evolutionary core of DP’s taxonomy, however, roots in the second-order mechanism rules. They subsume the entire set of rules that give impetus to change: rules for origination, rules for adoption, and rules for retention (Dopfer and Potts, 2007, p. 104). On an island, it is the trial-and-error strategy to find food that represents a process of origination. For these reasons, search strategies belong to second-order mechanism rules, rules that help, for instance, Friday come up with new ideas/opportunities that eventually lead him to the artifact of a coconut-picking rod. Having actualized all deductive steps to create the final operational rules of coconut “production”, Friday generated a technological artifact. He invented a new production technology and thus induced economic change. While Robinson still picks coconuts from the ground, Friday picks them from the trees. The onset of technological change roots in the change of rules on the generic level without yet having an impact on the operational level, that is, when the picking rod is eventually used for picking coconuts. Without second-order rules, the economic system would stall in an equilibrium, a static state of repetitive economic operations. This is why Dopfer (2006) distinguishes the operant from the generic level. It demarcates the static and repetitive dynamics from the Schumpeterian disruptive moment of economic change. In contrast to the operant level, on which economic actors perform operations on commodities and appear to behave in a continuous, routine-like behavior (conceived as optimal behavior in neoclassical eco­ nomics),1 the generic level defines the “deep level of ideas” – the source of novelty.

12.4

Generic rules and trajectories

DP’s taxonomy will certainly not allow accurate forecasts of the origination of rules, but deliver further insights in evolutionary economics. To a certain degree, novelty will probably remain a black box in human creativity. Nonetheless, it is an intriguing question how the homo sapiens oeconomicus generates novelty. Schumpeter (1934) conceived the creation of novelty as the creation of new combinations. In this regard, DP’s concept of the (re-)combination of generic rules is compatible with Schumpeter’s idea of new combinations. Furthermore, DP’s distinction between the generic and the operant level,2 explicitly fo­ cuses on the non-operational part of innovation that refers to cognition, that is, the creative process of (re-)combining generic rules. This very Schumpeterian idea of recombination as the source of innovation, DP put in the spotlight in their taxonomy by the conception of second-order rules. To give an example of how to flesh out the idea of origination and recombination of generic rules, a simple model as in Grebel (2009) may be helpful. It describes the creation of knowledge as a recombination of generic rules and the emergence of trajectories.3 Suppose we have a set of actors, each of which carrying a set of generic rules K = {k1, k2, …, kn} with ki ∈ R, which is restricted to the number of generic rules actors n, because they are assumed to be constrained in their cognitive capacity, i.e. their bounded rationality. To manage their daily lives, actors make use of certain combinations of generic rules which may prove viable or not to them and which repetitively lead to the same economic operations of the kind that

165

Thomas Grebel

generic rules

make Robinson repetitively “pick the windfall coconuts the shortest way”. These combi­ nations of generic rules construct operant-level rules (first-order rules). Friday, who harvests coconuts with a picking stick, also applies such a first-order rule. The invention of a picking rod, however, requires a new (re-)combination of generic rules. In this vein, let us assume Friday randomly selects p = 3 generic rules from the ones he carries. To consider the whole set of generic rules is infeasible to him as we assume him to be bound in his rationality. In a creative moment, he wants to generate new generic rules. When theorizing about this moment, the challenge is how to address the illuminating moment of successful invention, i.e. the creation of a potentially useful combination of generic rules. The stages of a creative process appear to be identifiable (Lubart, 2001), but the creative moment, due to epistemological reservations, is unpredictable because some­ thing new cannot be known in advance. Another challenge arises when it comes to mea­ suring knowledge. How to measure, for instance, the distance of knowledge? Generic rules cannot be properly quantified or compared on a cardinal scale. Putting this problem aside, let us simplify and assume generic rules as measurable by assigning a number to every generic rule. To mimic the moment of creativity, we add a random number to the generic rule that is most distant from the average of the three randomly chosen generic rules by Friday. This is the highly simplified version of Friday’s creative process of rule origination. While he integrates the new generic rule into his set of rules, he dismisses an existing generic rule of its previous set of generic rules. For instance, he forgets about the generic rule “pick coconuts from the ground”. Repeating this simple heuristic over and over again4 produces the evolution of Friday’s knowledge set as depicted in Figure 12.1. Thus, new generic rules – marked with circles – enter Friday’s knowledge set, where they remain until their removal when Friday forgets or substitutes them for other new generic rules. In the diagram, the persistence of generic rules in Friday’s knowledge set is indicated by gray lines. Because of the limited length (n) of his rule set, it solely is the scope of knowledge that expands, in other words, knowledge diversity increases.

time

Figure 12.1

Seceder model.

Note: Each circle indicates a newly generated generic rule, the lines in light gray track the persistence of the respective generic rule in Friday’s rule set.

166

Evolutionary modeling and the rule-based approach

With respect to DP’s taxonomy, this simple mechanism only illustrates one of three phases of meso-rule trajectories. The first phase, according to Dopfer (2006), pinpoints the moment of origination: Friday generates new generic rules in a creative moment, and the first time he applies the new rule to a real-world phenomenon, the phase of origination ends. Assuming a population greater than two on the island, in the second phase, the phase of adoption, a diffusion process would start during which potential carriers decide to adopt or not to adopt the new rule. If many do so, more and more meso-units emerge. Phase 3, the phase of retention, decides on the persistence of generic rules if carriers apply the rule repetitively. In this phase, it will show, whether the rule becomes institutionalized and thus part of actors’ behavioral routines in the sense of Nelson and Winter (1982).5 A further intriguing aspect revealed by the model is the emergence of bifurcating knowledge structures, which is due to the explicit assumptions of bounded rationality. Friday, unable to try out all possible combinations of generic rules at once, not having the capacity to store more than a given number of generic rules in his memory, he is not able to discover all possible generic rules of a given knowledge domain. Consequently, knowledge trajectories emerge. This emerging structural pattern produced by the model resembles the story of Technological Paradigms and Technological Trajectories by Dosi (1982), who argues that technological “[c]ontinuous changes are often related to progress along a technological trajectory defined by a technological paradigm, while discontinuities are associated with the emergence of a new paradigm” (Dosi, 1982, p. 147). In the model, the bifurcation towards a trajectory is accidental, as if Friday stumbles over a long stick that makes him think that he could use it to reach coconuts far up in the tree. Suppose that, beforehand, Friday had already invented a knife and a basket,6 the combination of these rules may have let him finally invent a picking rod. Once he considers the basic idea of a picking rod purposeful, further improvements of this new technology will only be incremental. He will only replace and improve parts of it. The model illustrates how simple it can be to apply DP’s rule approach. In general, it can be very helpful in agent-based modeling (Pyka and Grebel, 2006). With today’s computing capacity, modeling complex systems of rule-using agents has become quite easy. The fields of application are manifold. See, for instance, Castiglione (2020) in the field of financial markets, Paulin et al. (2018) in climate and energy topics, Castro et al. (2020), Dawid and Gatti (2018), or a survey by Revay and Cioffi-Revilla (2018).7 All these agent-based modeling approaches have in common that they implicitly follow the concept of a rulebased approach as they model agents applying specific behavioral rules.

12.5

Empirical quest

In general, little has been done in evolutionary economics to tailor a corresponding econ­ ometric approach for testing evolutionary concepts empirically. Foster and Potts (2009) argue that traditional econometrics such as multiple regression techniques seem to be of little help, because when structural change is prevalent many of the basic assumptions of traditional econometrics are violated. Therefore, they suggest a “(…) hybrid methodology of historical investigations, case studies, statistical analysis, econometric modeling and simulation/calibration” (Foster and Potts, 2009, p. 68). Whereas it is still an ongoing dis­ cussion to what extent existing econometric tools need adjustment, Blind and Pyka (2014, p. 1085) deliver a first constructive outline in the form of a “(…) methodological template 167

Thomas Grebel

for empirical research” to help operationalize the rule-based taxonomy of DP. They illus­ trate step by step how the main principles of DP’s approach can be applied within an evolutionary empirical analysis. As far as the model results in this contribution are concerned, it is an even more difficult task to operationalize them empirically. How to define, identify, and measure the generic level, i.e. the cognitive process of rule-based behavior? Only when actors actually articulate what rules they use in creating a particular artifact is there a chance of capturing the generic level in the spirit of DP. Patents can be a possible source of information on the use of generic rules. Each patent contains a combination of generic rules that suggest the creation of an artifact such as a “new product” or “technology”. One of the main benefits of this data source is that all patents contain detailed information about the knowledge classes they are assigned to. This classification system, the so-called international patent classification (IPC), repre­ sents the most fine-grained and comprehensive knowledge classification. Every patent is assigned to at least one IPC and thus indicates to which knowledge class the documented generic rules belong to. One of the most popular patent databases used by economists is the Patstat database provided by the European Patent Office (EPO). For its basic use see, for instance, De Rassenfosse et al. (2014), who also give further literature on patent analysis. Returning to Friday’s efforts to invent a tool to pick coconuts from the tree, the cor­ responding IPC code, where his new generic rules would be documented if he had the chance to apply for a patent on the island, would be A01D on the four-digit level. If he uses generic rules from other classes, this will also be reported in the patent document. The class A01D contains all patents about harvesting devices. For our purpose, we retrieved 253,755 patents containing this class. Aside from this class, these patents refer to additional 267 IPC out of possible 20,800 classes on the four-digit level.8 Figure 12.2 depicts the evolution of the body of knowledge in harvesting technology. The horizontal axis indicates the year of patent application proxied by the so-called priority date from the patent document, the vertical axis sorts IPC into canonical order. Note that this canonical order does not say anything about the actual distance of generic rules. Multiple references remain unconsidered. The first reference to an IPC is marked as a circle. As long as subsequent patents refer to the same four-digit IPC, the solid gray line indicates the persistence of that knowledge class. Hence, a new generic rule joins the body of knowledge on harvesting (circle) and persists as long as it is repetitively used (gray line) by successive patents. The first patent reported in patstat with IPC A01D, stems from 1877 (marked as circle in bold face). It was not before 1917, when inventor Summers Martin Vandver filed a patent suggesting the solution of how to gather cotton squares infested by boll weevils. In that patent, he cites IPC E01H, a class that belongs to the knowledge domain of con­ struction (bold face circle “Vandver”). In 1967, Bruce Wright invented a harvesting device by drawing on generic rules from the knowledge domain “electricity” (bold face circle “Wright”). Over time, more and more generic rules belonging to different knowledge domains joined the knowledge base on harvesting devices. Thus, the diversity of knowledge has continuously increased and new paradigms and trajectories have emerged. This simple empirical exercise is certainly not a full-fledged evidence for Friday’s theo­ retical example from above. A more elaborate inquiry would be required to reconcile caveats. To name a few, in Friday’s case, we have a single actor. In the empirical case, we deal with many different applications and inventors. The distance measure for knowledge is 168

International Patent Classification

Evolutionary modeling and the rule-based approach

H

ELECTRICITY

G

PHYSICS

F

MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING

E

FIXED CONSTRUCTIONS

D

TEXTILES; PAPER

C

CHEMISTRY; METALLURGY

B

PERFORMING OPERATIONS; TRANSPORTING

A

HUMAN NECESSITIES 1stharvestingpatent

Wright

Vandver

1880

Figure 12.2

1900

1920

1940

1960

1980

2000

Trajectory of harvesting patents.

strongly simplified. Nevertheless, the homology between the theoretical reasoning along the lines of DP and the corresponding empirical illustration remains promising for future research avenues.

12.6

Benefits of the taxonomy

The advantages of DP’s taxonomy are manifold. First and foremost, the distinction between the generic and the operant level allows a clear focus on the underlying mechanism of eco­ nomic evolution. From a purely operant-level perspective, i.e. neoclassical setting, economic actors would simply optimize their economic actions. Aspects of cognition, psychology, or sociology are ignored in order to simplify the complexity of human action. Conversely, this means that every observed phenomenon is conceived as a result of optimal behavior. In the case of real uncertainty, as in the case of a creative moment of invention or innovation, such a reverse conclusion seems unwarranted. It is inconceivable that individual actors or firms would be able to grasp all existing generic rules and combine them optimally in the sense of an unpredictable outcome. The benefits of an invention cannot be fully assessed exante. The consequences for the ontological framework of basic assumptions are straightforward: true uncertainty must lead to heterogeneous (Kirman, 2006), path-dependent (David, 2001), and fallible (Loasby, 2002) outcomes. In DP’s taxonomy, these basic assumptions are easy to implement. As generic rules are bimodal, or as DP put it, as there is “oneness” and “mani­ ness”, their actualizations can be manifold. Outcomes will be heterogeneous because eco­ nomic behavior is associated with a given set of generic rules an actor carries. Hence, behavior will be path-dependent to the extent to which the set of rules is persistent and the deductive procedure in decision making becomes a routine (Nelson and Winter, 1982). 169

Thomas Grebel

This leads directly to the second advantage. DP’s approach is more general and least restrictive in terms of choice of methodology. For instance, it does not reject methodological individualism, because the economic actor is solely a carrier of rules. Under certain cir­ cumstances, these rules may be sufficient for optimal behavior. It also allows us to think in terms of populations, since a meso-unit is the population of a generic rule. In contrast to other heterodox approaches such as “Generalized Darwinisim” (Hodgson, 2013) or “Organizational Ecology” (Dollimore, 2014) that seem to be absorbed in arguing about the interpretative sovereignty of the congruence of analogies between the Darwinian “evolu­ tionary” concept and its counterparts in the realm of economics (Thomas, 2018), the focus on “rules”, as the smallest element or building block of human behavior, helps prevent researchers from getting distracted by sidetracks in his or her economic analysis. In DP’s taxonomy, rules define the domain of human actions, irrespective of whether individual behavior is directed by immutable routines (instinctive behavior), optimal behavior (per­ fectly rational behavior), or contextual (adaptive behavior) factors. By contrast, the attempt to impose biological analogies on economic behavior neglects the generic aspect of human action. Imposing biological analogies as prescriptions for individual behavior neglects the generic inherent dynamics of economic decision making by a Homo sapiens oeconomicus. In this respect, DP’s approach is more general because it does not take sides. Rule carriers, i.e. individual economic actors, remain decision makers, but can also be carriers of populationbased generic rules. Moreover, this approach easily blends in with other disciplines such as psychology or sociology and is also compatible with behavioral economics. Last but not least, a benefit worth mentioning is the small step from theorizing towards empirical implementation. Theorizing about the link between generic rules and their actu­ alization is tantamount to hypothesizing about cause and effect of human behavior. Identifying rules and their interplay suggest an immediate design for empirical testing as Blind and Pyka (2014) demonstrate.

12.7

Conclusion

This contribution is meant to motivate rule-based thinking in the sense of the DP approach. It can facilitate the analysis of evolutionary phenomena that go beyond the operant level. To provide a complete picture of the richness of Dopfer and Potts’ taxonomy is far beyond the scope of this contribution. It was introduced very recently so its full potential has not yet been fully exploited. Many questions still need to be answered (Runde, 2009), some of which have already been answered (Dopfer and Potts, 2010), and others yet need to be clarified. Nonetheless, its completeness and consistency have been acknowledged by various scholars such as Ostrom and Basurto (2011), Beinhocker (2011), and Strohmaier (2010). Some merits have been addressed in this contribution. Putting the focus on “rules” offers a flexible, versatile set of instruments to engage in evolutionary analysis. The taxonomy is neither contradictory to the maximization hypothesis nor to the metaphor of a state of equilibrium. Hence, it does not engage in futile methodological discussions (Boland, 1981) – maximization simply represents a special case of rule-using economic actors. Group or population thinking can be modeled via zeroth-order rules as they constitute the set of socially approved and persistent (generic) rules. Routinized behavior arises from first-order rules, while their bimodal actualizations across carriers lead to heterogeneity. With the approach by Dopfer and Potts, historicity, path dependence, bounded rationality, and heterogeneity become a necessary features in modeling economic behavior, and last but not 170

Evolutionary modeling and the rule-based approach

least, the focus of evolutionary economic modeling is directly put on economic change due to the concept of second-order rules and the constantly ongoing process of rule origination, selective adoption, and retention for repeated operations.

12.8 Classification of rules The following table illustrates the rule types and classes of DP’s taxonomy. Table 12.1 Orders and Classes of Generic Rules Own compilation, Compare Dopfer and Potts (2007) Rule types and classes Order of rules

Example

Class of rule

zeroth-order rules

try to survive windfall coconuts belong to Friday trial-and-error searching behavior to find the food location creation of a new rule, i.e. invention and adoption of the coconut-picking rod the optimizing behavior of following the shortest route to the palm trees the steps of applying the invented rod to pick coconuts

Subject-cognitive Object-social Subject-behavioral Object-technical

second-order rules first-order rules

Subject-behavioral Object-technical

Table 12.2 summarizes the corresponding classes of generic rules by Dopfer and Potts (2007). For a short summary, see also Blind (2017). Table 12.2 Classes of Generic Rules Own Compilation, See Also Dopfer and Potts (2009) and Blind (2017). Generic Rule Classes Subject rules Cognitive

Object rules

Behavioral

e.g. to survive; to win the e.g. to find food; use coconut competition trial-and-error searching behavior

12.9

Social

Technical

e.g. windfall coconuts belong to Friday

e.g. how the coconutpicking rod should work

Biography

Thomas Grebel is a full professor of economics at the Technische Universität Ilmenau in Germany. He is the head of the economic policy group. In his research, he focuses on innovation economics, economic modeling, and applied econometrics.

Notes 1 See Dopfer and Potts (2009) for more details concerning the distinction between a neoclassical perspective on the operant level and the evolutionary perspective on the generic level. 2 The operant level refers to the level at which economic operations or artifacts are observable. 3 The model builds on the basic concept by Dittrich et al. (2001) and Dittrich and Banzhaf (2001).

171

Thomas Grebel 4 See Grebel (2009) for details. 5 Compare Grebel (2009). 6 Think of a knife as an actualization of the generic rule “part a thing with a sharp object” and of a basket: “use an object to contain things in it”. 7 For further inspiration see the Handbook of Computational Economics. 8 The International Patent Classification comprises 8 sections (first digit of the IPC symbol ranging from A to H), 10 classes (second digit ranging from 0 to 9), 10 subclasses (third digit ranging from 0 to 9), and 26 groups (fourth digit ranging from A to Z). Successive subgroups are neglected.

References Barreto, H. (1989). The Entrepreneur in Microeconomic Theory. Routledge, London and New York. Beinhocker, E. D. (2011). Evolution as computation: Integrating self-organization with generalized Darwinism. Journal of Institutional Economics, 7(3):393–423. Blind, G. and Pyka, A. (2014). The rule approach in evolutionary economics: A methodological template for empirical research. Journal of Evolutionary Economics, 24(5):1085–1105. Blind, G. D. (2017). The rule-based approach in the analysis of economic change. In The Entrepreneur in Rule-Based Economics, pp. 13–20. Springer. Boland, L. A. (1981). On the futility of criticizing the neoclassical maximization hypothesis. The American Economic Review, 71(5):1031–1036. Castiglione, F. (2020). Agent-based modeling and simulation, Introduction to Complex Social and Behavioral Systems: Game Theory and Agent-Based Models, pp. 661–665. Castro, J., Drews, S., Exadaktylos, F., Foramitti, J., Klein, F., Konc, T., Savin, I., and van den Bergh, J. (2020). A review of agent-based modeling of climate-energy policy. Wiley Interdisciplinary Reviews: Climate Change, 11(4):e647. David, P. A. (2001). Path dependence, its critics and the quest for ‘historical economics’. Evolution and Path Dependence in Economic Ideas: Past and Present, 15:40. Dawid, H. and Gatti, D. D. (2018). Agent-based macroeconomics. Handbook of Computational Economics, 4:63–156. De Rassenfosse, G., Dernis, H., and Boedt, G. (2014). An introduction to the Patstat Database with example queries. Australian Economic Review, 47(3):395–408. Dittrich, P. and Banzhaf, W. (2001). Survival of the unfittest? - The Seceder model and its fitness landscape. In Kelemen, J. and Sosik, P., editors, Advances in Artificial Life, pp. 100–109, Berlin. Springer. 6th European Conference on Artificial Life, Prague, September 10–14, 2001. Dittrich, P., Ziegler, J., and Banzhaf, W. (2001). Artificial chemistries – Review. Artificial Life, (7):225–275. Dollimore, D. E. (2014). Untangling the conceptual issues raised in Reydon and Scholz’s critique of organizational ecology and Darwinian populations. Philosophy of the Social Sciences, 44(3):282–315. Dopfer, K. (2006). The Origins of Meso-Economics – Schumpeter’s Legacy. Technical Report 0610, Max Planck Institute of Economics, Evolutionary Economics Group. Dopfer, K. and Potts, J. (2007). The General Theory of Economic Evolution. Routledge. Dopfer, K. and Potts, J. (2009). On the theory of economic evolution. Evolutionary and Institutional Economics Review, 6(1):23–44. Dopfer, K. and Potts, J. (2010). Why evolutionary realism underpins evolutionary economic analysis and theory: A reply to Runde’s critique. Journal of Institutional Economics, 6(3):401–413. Dosi, G. (1982). Technological paradigms and technological trajectories: A suggested interpretation of the determinants and directions of technical change. Research Policy, 11(3):147–162. Foster, J. and Potts, J. (2009). A micro-meso-macro perspective on the methodology of evolutionary economics: Integrating history, simulation and econometrics. Schumpeterian Perspectives on Innovation, Competition and Growth, pp. 53–68. Springer. Grapard, U. (1995). Robinson crusoe: The quintessential economic man? Feminist Economics, 1(1):33–52. Grebel, T. (2004). Entrepreneurship – A New Perspective. Routledge, London and New York. Grebel, T. (2009). Technological change: A microeconomic approach to the creation of knowledge. Structural Change and Economic Dynamics, 20(4):301–312.

172

Evolutionary modeling and the rule-based approach Hodgson, G. M. (1998). The approach of institutional economics. Journal of Economic Literature, 36(1):166 – 192. Hodgson, G. M. (2013). Understanding organizational evolution: Toward a research agenda using generalized Darwinism. Organization Studies, 34(7):973–992. Kirman, A. (2006). Heterogeneity in economics. Journal of Economic Interaction and Coordination, 1(1):89–117. Loasby, B. J. (2002). Knowledge, Institutions and Evolution in Economics. Number 2. Psychology Press. Lubart, T. I. (2001). Models of the creative process: Past, present and future. Creativity Research Journal, 13(3–4):295–308. Marshall, A. (1948). Principles of Economics. MacMillan and Co., London, 8th edition. First pub­ lished 1920. Nelson, R. R. and Winter, S. G. (1982). An Evolutionary Theory of Economic Change. Cambridge University Press, Cambridge, MA. Ostrom, E. and Basurto, X. (2011). Crafting analytical tools to study institutional change. Journal of Institutional Economics, 7(3):317–343. Paulin, J., Calinescu, A., and Wooldridge, M. (2018). Agent-based modeling for complex financial systems. IEEE Intelligent Systems, 33(2):74–82. Pyka, A. and Grebel, T. (2006). Agent-based modelling—a methodology for the analysis of qualitative development processes. In Agent-Based Computational Modelling, pp. 17–35. Springer. Revay, P. and Cioffi-Revilla, C. (2018). Survey of evolutionary computation methods in social agentbased modeling studies. Journal of Computational Social Science, 1(1):115–146. Runde, J. (2009). Ontology and the foundations of evolutionary economic theory: On Dopfer and Potts’ general theory of economic evolution. Journal of Institutional Economics, 5(3):361–378. Schumpeter, J. A. (1934). The Theory of Economic Development: An Inquiry into Profits, Capital, Credit, Interest and the Business Cycle, trans. R. Opie in 1934. Harvard University Press, Cambridge, MA. Schumpeter, J. A. (1942). Capitalism, Socialism and Democracy. Harper & Row. Shionoya, Y. (1998). Schumpeterian evolutionism. In Davis, J. B., Hands, D. W., and Maki, U., (eds.), Handbook of Economic Methodology, pp. 436–439. Edward Elgar, Cheltenham. Strohmaier, R. M. (2010). The general theory of economic evolution. Thomas, R. (2018). The claims of Generalized Darwinism. Philosophy of Management, 17(2):149–167. Veblen, T. (1898). Why is economics not an evolutionary science? The Quarterly Journal of Economics, 12:373–397. von Mises, L. (1959). Human Action. William Hodge, London.

173

13 CONTINGENCY IN EVOLUTIONARY ECONOMICS Causality and comparative analysis Marco Lehmann-Waffenschmidt Thomas Grebel 13.1

Comparative evolutionary analysis

Economic activities take place as processes in real time. Consequently, it is necessary to develop and apply process oriented rather than static patterns of thought and explanations for an adequate analysis. Following these lines, we will develop further the approach of comparative economic analysis from the traditional static and dynamic version and will arrive at a “comparative evolutionary analysis”. What this means is obvious. We start from an – at least partially – “open loop” evolution of an economic system as a point of reference and compare it with other “plausible“ evolutions of the same economic system with respect to the relevant variables. “Open loop” means a process which is at least partially open at each moment in time, i.e. which is not completely determined in its course and its final state (Witt 2004, Cordes 2006). “Plausible” means justifiable by a scientifically well-founded reasoning. In an ex-post analysis of an evolving real economic system, the factual evolution of the system is the reference evolution, while the comparative evolutions are constructed by thought experiments and consequently are counterfactual. The methodologically crucial question is how the evolutions to be compared can be reasonably constructed. In the case of comparative static and comparative dynamic analysis, the construction of alternative var­ iants to be compared with the reference case is not problematic – model parameters are changed, or different stochastic realizations are considered in a stochastic-dynamic modeling. But it remains to be clarified in the following how alternative evolutions to be compared can be modeled in a systematically sound way for a comparative evolutionary analysis. It is the first aim of this contribution to give an answer to this question by means of the contingency analytic approach. But what about the second issue in the title of this contribution – causality? As we will see the contingency analytic approach also lays the foundations to develop an intuitive method to measure the causality relationship of any two states at different times of a process. To be more precise, the contingency analysis method in a natural allows for a gradual measurement of causality which obviously corresponds to the multiple alternatively possible evolutions used by the contingency concept. The core element of the contingency analytic approach is the conviction that economic evolutions, or processes, in real time are at least partially open loop as had been mentioned

174

DOI: 10.4324/9780429398971-15

Contingency in evolutionary economics

above so that they can reach “multifurcation points” during elapsing time where multiple different continuations are possible. This can be the case first in processes imagined into the future – forecasted or planned – so that different future “scenarios” may emerge, and second in the historical retrospective. Since in the retrospective naturally, there can always be only one realization of the factual course of the process, all other evolutions that can be constructed as alternatively possible in the past become counterfactual alternatives. Examples of multifurcation points in time at which other decisions than the factual ones would also have been possible can be found in the history of IBM (missed entry into PC production at the end of the 1980s), VW (emissions measurement manipulations on a large scale until 2015), Kodak (missed entry into digital photography in the 1990s), or Bayer (purchase of Monsanto in 2018).1 Systematic thinking in possibilities already had its origins in ancient Greek philosophy, more precisely in the modal logic of Aristotle (4th century B.C.). The modern concept of a contingent event as “not necessary, but not impossible” goes back to Gottfried W. Leibniz (1646–1716) (Latin contingere = to happen). In the epistemological debate of our time, the names Kripke, Lewis, Lübbe, Luhmann, Marquard, and Rorty are associated with the contingency approach, and there are also striking emanations of contingency thinking into different academic fields such as history, religion, politics, literature, economics, sociology, and statistics.2 The aim of this contribution is not to trace the development of the history of ideas and theories or to portray the current state of the epistemological debates on the concept of contingency, but to clarify the significance and role of a contingency-oriented approach for evolutionary economics and to provide the reader with the announced ana­ lytical contingency concept.3 The ultimate purpose is to contribute to improve forecasts and recommendations and instructions to control and shape real economic processes. Our life is essentially shaped by contingencies as is already shown by every subjunctive (“If only he had …”, “What if … ?”, “I could go now”), by a trial after a traffic accident, or by the search for those agents responsible for an undesirable economic development caused by a political decisions. Phileas Fogg’s round-the-world trip in 80 days in Jules Verne’s novel, every chess match or soccer game, or a panel discussion on a controversial topic are (at least par­ tially) open loop processes in terms of course and outcome, in which at several points in time contingencies appear to be possible. In other words, their courses are neither compelling, nor arbitrary, and they can be influenced by path dependencies which means restrictions for the future coming from the past. Particularly, biographies often show multifurcation points. The poet Jean Paul has set a literary monument to this idea as early as 1818 in his autobiographical “Konjekturalbiographie”, Max Frisch and Yasmin Reza show in a dramatic way alternative possible life courses in their stage plays “Biografie. A Game” (1967 by Frisch) and “Trois Versions de la Vie” (2000 by Reza). And prominent biographies in the past and present provide striking examples – Alexander the Great almost fell victim to a serious illness as a child, Otto von Bismarck was rescued at the last moment in a swimming accident in the Atlantic Ocean in the early 1860s, and Friedrich Ebert, the first president of the German Weimar Republic after World War I, was not an all predestined for this position since he was born as the son of a tailor and learned the saddlery trade as a youth. Movies like Look Both Ways, Back to the Future, Sliding Doors, Groundhog’s Day, and Frank Capra‘s Isn’t life beautiful? from 1946, or Kurosawa’s Rashomon from 1950 play with the idea of contingency and counterfactual thinking in an inspiring way. A natural question is whether the choice of one alternative in a multifurcation situation is ruled by chance. In fact, this may be the case – the movie Sliding Doors illustrates this when 175

Thomas Grebel

the principal actress Gwyneth Paltrow reaches the tube before it starts in variant 1, or she does not (the doors are closed) in variant 2. However, in many cases this is not the case, i.e. the choice for one possible alternative is not a matter of random influences: The figure of Phil Connors in Groundhog’s Day deliberately tries out different behavioral patterns in (involuntary) repetitive loops, in Isn’t Life Beautiful? the protagonist George Bailey ima­ gines counterfactually how his life would have developed in the past without him, and in Rashomon a 12th-century murder of a samurai is reconstructed in four different, convicting variations. Documentary TV-series on counterfactual historical analysis (“What if …”) show the interest of historians and the public, too, in using this method to understand and evaluate processes retrospectively, or to estimate them in advance with regard to the responsibilities and performance qualities of the actors. It is interesting that the number of reasonable alternative processes in case studies remains as experience shows in the (small) single-digit range, i.e. serious counterfactual thought experiments do not lead to an unmanageable jumble of overflowing hypothetical fantasy worlds, as one might fear. The omnipresence of economic thinking in contingencies can be seen, for example, in a simple market diagram with a demand and a supply function in a price-quantity coordinate system. In fact, the two curves describe contingent plans of the modeled consumers and suppliers. Clearly at any time, one will always observe a unique price-quantity combination in the positive orthant of the price-quantity coordinate system. All other points on the supply or demand curves are counterfactual alternatives to this situation. Another example is the economic key concept of opportunity costs. Here, the expected, or realized, net income of a certain activity is compared to the expected, or realized, net income of alter­ native activities, which would also be possible for the agent. The highest alternative net revenue that can be generated means the opportunity costs of the activity chosen. Thus opportunity costs are a contingent concept.

13.2

The graphical-analytical contingency concept 4

The basic idea is to model the “possibility environment” of a factual evolution of an eco­ nomic system, e.g. a national economy, an industry, or a company, in real historical time in an appropriate graphical way. This way to model contingency characteristics of a process should make well-founded conclusions possible on the causality relations between any two states at different points in time of the analyzed process.5 This may become particularly relevant in the context of desired, or undesired process characteristics since it allows ex-post responsibilities to be determined and the performance to be evaluated and ex-ante design measures to be assessed with regard to their suitability for certain goals.6 The representation of the possibility environment of a process by a contingency graph suggests that the causal relationship intensity between any two states of the process at different points in time is from its nature rather gradually (valued between 0 and 1) than binarily characterized (“causal related” or “not causal related”). How the causal relationship between two states of a process at different times can gradually be measured will be shown in the next section. The graphical-analytical contingency concept can be represented graphically by a “dia­ chronic contingency di-graph”, the “contingency graph” modeling the contingency struc­ ture of a process. The process π in Figure 13.1 below defined by the node sequence (E1, E2I, E3VI) and its possibility environment (= all other edges and nodes of the graph of Figure 13.1) illustrates this. As a di-graph, the contingency graph consists of two types of elements: Of nodes, which represent states or events of the modeled process and are 176

Contingency in evolutionary economics

Figure 13.1

A contingency graph with 3 points in time, 11 nodes (1 initial node, 4 inner nodes at date t2, and 6 end nodes), 14 edges, and 4 cycles. π denotes the factual reference path.

diachronically marked with time indices, and of edges, which represent at each node the possibilities of how the process can proceed in the next time step. If there is only one edge at a certain node, then the state of the process at the next point in time is determined. If there are two edges, the node represents a bifurcation point – think of “yes” or “no” or of “right” or “left” – and if there are more than two nodes, it is a multifurcation point.7 The historical time is the dimension of the x-axis of the coordinate system. The ordinate axis symbolizes the space of possible system states that can be reached during the process. To be sure, the state space may not only be symbolized, but may be represented accurately by the scalar one-dimensional coordinate axis as in the case of the GDP of an economy. Some questions may arise: How do we know how to construct the contingency graph, i.e. at which points in time are which alternatively possible states relevant? In particular, how must the “time granularity of the horizontal time axis” be designed, i.e. which points on the time axis are relevant as time coordinates of nodes of a contingency graph? Furthermore: Which dimension does the ordinate have, and which edges have to be drawn as contingent process possibilities at the nodes in order to adequately represent the true possibility en­ vironment of the factual process? And last but not least: Does modeling the possibility environment of real processes in economics not inevitably result in unmanageably complex contingency graphs that are as useful for further analysis as a “1–1 city map” of Paris – i.e. Paris itself – for a tourist as orientation device? The basic answer to all these questions is that a contingency graph like any modeling of a real phenomenon can only represent the existing expert knowledge, here about the process under consideration and its contingent alternatives. Of course, objective, or subjective knowledge deficits cannot be overcome by the contingency conceptualization per se. The question of the manageability of contingency graphs in real applications cannot be answered in general abstract terms – the complexity of the contingency graph of a real modeled process depends on its specific nature and on the present state of knowledge about its possible environment. Therefore, an answer to this question can only be given by means of empirical findings. Indeed, the experience in the application of the analytical contingency concept to (economic) historical case studies shows that the contingency graph structure is 177

Thomas Grebel

Figure 13.2

An example of a convergent “equifinal” contingency graph.

by no means prone to over-complexity. Instead, the number of relevant points on the time axis and the nodes of the graph are manageable, and the number of edges at the nodes usually remains in the small single-digit range. For example, the counterfactual cliometric8 studies by Robert Fogel to investigate whether the de facto railroad expansion in the USA in the 19th century was necessary for the economic take-off of the USA in terms of valueadded performance contain only one counterfactual:9 There is no railroad network ex­ pansion in the USA, but canals and land routes get expanded. Indeed, Fogel and his group found out that the railroad expansion was not a necessary precondition for the growth and prosperity of the US economy in the 19th century. Also a contingency analysis of the question of why light water and not gas-graphite reactor technology has prevailed world­ wide in the peaceful use of nuclear power after 1945 or case studies of historically “lockedin” suboptimal technologies (VHS video cassettes, “QWERTY” typewriter letter systems, etc.) show clearly structured and manageable contingency graphs. The contingency graph in Figure 13.2 below shows the special case of equifinality where two disjunct paths in a contingency graph finally converge and end in the same final state Ei+2, which, as in Figure 13.2, may then possibly be stationary stable. To be sure, even if equifinality may occur in reality, this does not mean the irrelevance of a contingency analysis, since in reality, it is crucial what happens in the – possibly long – period before ti+2 and, moreover, the equifinality, i.e. the convergence in ti+2, is (more or less) uncertain before ti+2. Equifinality is a special – and extreme – case of the property of directedness of a process. A drift is a special case of non-directedness in that it appears to be directed, but one cannot conclusively justify or explain it. A desired directionality of a process in the sense of progress and development is referred to as “orthogenesis” or “anagenesis”. “Teleological” processes have a clear direction to a certain goal, and “teleonomial” processes are subject to laws of progression, but nevertheless have degrees of freedom for a partial openness of course and final state. But the analytical contingency concept can be even more. For example, path depen­ dencies can be integrated into the contingency graph, which is illustrated by the following Figure 13.3: How to proceed further from state E3III essentially depends on where you come from reaching E3III. And the concept of contingency offers the possibility to distinguish between “pro­ grade” and “retrograde alternative sets” of a node Ei at the time ti, i.e. the “prograde” 178

Contingency in evolutionary economics

Figure 13.3

Path dependency in the contingency graph.

nodes in the graph, which are connected by an edge to Ei in the next time step ti +1, i.e. which can be reached from Ei, and the “retrograde” alternative set of those nodes in the graph at the previous time ti-1, from which an edge to Ei exists in the graph. Both sets of alternatives can be empty in principle, but they are not, if one postulates that the graph must not contain nodes which are newly created (so that the retrograde set of alternatives would be empty), or where a path in the contingency graph ends (so that the prograde set of alternatives would be empty). How are the contingent choices at multifurcation points, i.e. the selection of a certain edge, realized in real processes? One can distinguish between two modes – on the one hand, the mode of a “decision-determined, or situational regime” and, on the other hand, the mode of a “structural, or systemic regime”. A decision-determined, or situational, regime exists when a person or a group makes a decision and has degrees of freedom so that several possible outcomes can result. This type of regime is typical for the history of companies – as we had seen before the automobile manufacturers would very well have been able to avoid the fraud in the emission measurement of their vehicles, or Bayer could have refrained from acquiring Monsanto. A structural or systemic regime existed, for example, in 18th-century Europe when, due to several structural conditions, England realized the (first) Industrial Revolution before France.

13.3 The gradual measurement of causality relationships between different states of a process by the degree of prograde causality and the degree of retrograde causality The contingency analysis approach allows for the conceptualization of two degrees of causality – the prograde and the retrograde degree of causality. While the prograde degree of causality measures the degree of the causation of a later state Ei+n of the graph at time ti+n by an earlier state Ei at time ti by a real number between 0 and 1, the retrograde causality between Ei+n and Ei is measured by a real number between 0 and 1, which determines the degree of the former state Ei as the cause of the later state Ei+n. To be more specific, the n-prograde degree of causality between Ei and Ei+n is calculated by dividing the number of paths in the graph leading from Ei to Ei+n by the number of paths leading from Ei to any state of the graph at time ti+n. Obviously, this quotient must be between 0 and 1, and 0 and 1 are included. Accordingly, the m-retrograde degree of causality 179

Thomas Grebel

Figure 13.4

Retrograde alternative set with four nodes.

Figure 13.5

Prograde alternative set with five nodes.

between Ei+m and Ei is calculated by dividing the number of paths in the graph leading from Ei to Ei+m by the number of paths leading from any state at time ti to Ei+m. Obviously, also this quotient is always from the closed unit interval [0, 1] and both quotients are different in general. The following two graphs show two simple and instructive special cases with n = m = = 1, i.e. with only a one-time step: The 1-retrograde causality degree between Ei and any node Ei-1k in Figure 13.4 is ¼, the 1-prograde of Ei and Ei+1j in Figure 13.5 is 1/5. Another example: In Figure 13.1 above, the 2-prograde causality degree between the initial state E1 and E3III at time t3 is 3/11, since there are 3 paths from E1 to E3III and 11 paths from E1 to 1 of the 6 possible states at time t3 which leads to 3/11.

180

Contingency in evolutionary economics

13.4

On the relationship between the contingency approach and the causal-logical terms “necessary” and “sufficient”

The question may arise what the relationship is between the classical causal-logical cate­ gories necessary, sufficient, and equivalent of cause-and-effect relationships on one hand and the contingency approach on the other. Could the concept of contingency possibly be described completely with the help of these classical logical cause-and-effect categories which would mean that the contingency approach is merely a reformulation of what is already known? A basic argument will show that this is by no means the case. The contingency approach, when applied to a socio-economic system, models the “possibility environment” of the factual states of a historical – or the expected states of the future – evolutionary path of the system. The causal-logical terms “necessary” and “suffi­ cient”, on the other hand, refer abstractly and generally to possible causal-logical relations in which two elements A and B can relate to each other with respect to the way B is caused by A. But to be sure, there are partial touchpoints of both approaches which, however, cannot be presented here.

13.5

The extension of contingency analysis by probabilities – On the relationship between the causal contingency analysis approach and probability theory

Due to its construction, the contingency approach is fundamentally different from the stochastic, or probabilistic approach. While the stochastic approach is based on random experiments in populations, or on subjective probability attributions by agents, the con­ tingency approach does not necessarily need any probability attributions. The analytical contingency modeling is diachronic-time related and can therefore take path dependencies into account, is process-specific in content and subject matter, so that comprehensible and plausible justifications can be given for all states and paths of a modeled process, and can differentiate between forward looking progrades and in time backward looking retrograde causal relationships by the time reference in contrast to the stochastic approach. Thus, the contingency approach not only allows for statements about correlations, but particularly leads to conclusions about causality relations between diachronic states in both prograde and retrograde time perspective. In particular, this reduces the risk of “post-hoc ergopropter-hoc” errors. On the other hand, despite the systematic differences, it is obvious to extend the concept of contingency by probabilities, if probability attributions are available for the edges of the contingency graph – be they objective or subjective. Graphically, the probability extension in the model of a contingency graph is achieved by weighting all edges of the corresponding prograde set of alternatives at each state or node of the graph according to a probability distribution with probabilities, so that they add up to 1. The formal-analytical basic element for this probability-enhanced contingency analysis is the path probability weight, or simply path weight, of any path in the graph, which is defined in an obvious way as the conventional conditional probability of this path in the graph, i.e. as the product of the probabilities of the edges constituting the path under consideration. If one now wants to extend the concepts of the prograde and retrograde degree of causality by probabilities, it is obvious to replace the number of connecting paths in the dividends and divisors of both causality degree quotients by the sum of the path (proba­ bility) weights of the same connecting paths respectively.

181

Thomas Grebel

Figure 13.6

A contingency graph with probability weights.

A question remains – could it be the case that the probability-enhanced and the basic contingency-causality concepts could be derived by suitable reformulations from the con­ ventional probabilistic approach and are thus obsolete? To be sure, this suspicion proves to be false, as simple counterexamples show. For instance, the path (probability) weight and the prograde probability-extended degree of causality between two states Ei to Ei+n cor­ respond exactly to the conventional conditional probability. But here the direct corre­ spondence between conditional probability and the presented causality concepts stops as the following example of Figure 13.6 shows. The retrograde probability-weighted causality between the two states E3IV and E2II is 2/7, if one assigns a probability equal distribution at the states, or nodes, E2I and E2II respec­ tively to the outgoing edges, i.e. the probability ½ for the edges 5 and 6 and ¼ for the edges 1 to 4. This value has nothing to do with the conditional probability with the value 1/2 to reach the state E3IV from E2II, or with the not probability-weighted retrograde degree of causality which is 1/3. The conditional probability of E1 reaching state E3VI is like the probability-weighted degree of prograde causality 1/6 × 1/3 = 1/18. But the not probabilityweighted prograde degree of causality is 1/10, and if an equal probability distribution of the outgoing edges would be assumed at all nodes or states, the conditional probability to reach the state E3VI from E1 would still not have the value 1/10 of the not probability-weighted prograde degree of causality, but the value 1/12.

13.6

Conclusions – What are the merits of the contingency analysis approach from the perspective of evolutionary economics?

Contingency is a distinctive characteristic of our process-dominated world. Taking this seriously one has to construct ex-post alternative possible processes for those factual pro­ cesses which are – at least partially – open with respect to their course and to their final state, i.e. which are neither completely determined nor stochastic, arbitrary, or erratic. Thus the factual and counterfactual evolutions in a contingency analysis are not simply attributed 182

Contingency in evolutionary economics

to random or arbitrarily erratic influences, but can be characterized by comprehensible and conclusively founded explanations. Contingent – i.e. posssible, but not necessary – en­ vironments of factual processes are created. For future-oriented processes, the same pro­ cedure applies in principle, but the factual reference process naturally does not yet exist. Consequently, the graphical-analytical contingency approach has been presented here in its two variants – first, the ex-post variant using the counterfactual analysis, and second, the ex-ante variant using the scenario analysis. To be sure, neither variant is based on wishful fantasies or arbitrary speculations. This means in particular that in its ex-post form, the contingency analysis approach does not aim for a “better” counterfactual past, but for a better understanding of factual processes with respect to the causal structure between diachronically apart events of a process. The contingency approach contributes to the investigation of the existence of historical regularities in the course of real processes. It allows possible causal relationships between states at different points in real time during the course of a process to be investigated and their intensities to be determined gradually. Such historical laws of course can be recurrent patterns, i.e. structurally recurring sections of course over the course of time, which first allow empirical generalizations and finally have to be explained theoretically. States of a process with an arbitrary irregular – and thus unpredictable – course, on the other hand, are singular events that are only accessible to case-by-case casuistic analysis, but cannot be explained scientifically in a systematic way and thus are not really capable of being theo­ rized. Well-founded predictions and recommendations for action with regard to certain well-founded goals are thus not possible, especially in the economic sector. But these are exactly what is expected and needed by a socially relevant economic science.

Notes 1 In fact, these examples are even “bifurcation points” because only two alternatives were possible in each case – the de facto decision and the opposite: Bayer and VW could have simply ceased their activities, and IBM and Kodak could each have invested in the new technology environment that had already emerged. 2 In the science of history, the approach of so-called counterfactual, virtual, alternative, or parallel history naturally plays an important role. Even ancient Greek and Latin historians like Thukydides or Tacitus have used counterfactual argumentations, and Toynbee and Churchill also wrote counterfactual studies. In current historical research, the historian Alexander Demandt in particular has rendered outstanding services to this approach. Ortmann (1995) has introduced the concept of contingency from a business management and corporate sociology perspective into the economic debate. The paleontologist Stephen J. Gould made the contingency concept prominent in evolu­ tionary biology in the late 1970s ( Lehmann-Waffenschmidt 2018). 3 What is not meant by this is the crude idea of “alternative facts” as it is sometimes used in our days because this is obviously nonsense since the term “facts” linguistically unambiguously designates something that really exists and not something that does not exist - be it possible or impossible, desired or undesired. 4 In this contribution, mainly the graphical representation is used. For the formal-analytical exact definitions and derivations, the reader is referred to Lehmann-Waffenschmidt 2010, 2022. 5 Hodgson 2004 has also dealt with causality in the context of evolution, but in a different context. 6 The question of which process courses are considered desirable by which subjects does not fall within the focus of this approach. 7 The contingency graph must actually be called “graph” and not “tree”, because a contingency graph can have converging edges during elapsing time, i.e. convergent subprocesses that generate cycles in the graph - which is normally not possible with a tree. Dopfer (2001) has presented an analogous “tree” with convergent edges and cycles of a contingency graph.

183

Thomas Grebel 8 “Kliometrics” refers to the methodical approach of analyzing historical processes using econometric methods. Klio is the name of the ancient Greek muse of historiography (originally spelled in ancient Greek with a “K”). R. Fogel received the Nobel Memorial Prize for economics in 1993. 9 Kliometric counterfactual analysis usually only assumes one counterfactual that is historically plausible, i.e. that could have been realized in principle. The contingency approach presented here, on the other hand, allows in principle for several counterfactuals in one state, which would have been historically feasible in principle.

References Cordes, C. (2006) Darwinism in Economics: From Analogy to Continuity. Journal of Evolutionary Economics, pp. 529–541. Dopfer, K. (2001) History-Friendly Theories in Economics: Reconciling Universality and Context in Evolutionary Analysis, in J. Foster and J. S. Metalfe (eds.), Frontiers of Evolutionary Economics. Competition, Self-Organization and Innovation Policy, Cheltenham: Edward Elgar. Hodgson, G. M. (2004) Darwinism, Causality and the Social Sciences. Journal of Economic Methodology, 11(2), pp. 175–194. Lehmann-Waffenschmidt, M. (2010) Contingency and Causality in Economic Processes – Conceptualizations, Formalizations, and Applications in Counterfactual Analysis. European Review, 18(4), 481–505. Lehmann-Waffenschmidt, M. (2018) Counterfactuality, Contingency, Evolution and Co. A Counterfactual Conversation, in C. Meißelbach, J. Lempp, and S. Dreischer (eds.), Perspectives from Science and Society. Springer Verlag, pp. 279–302. Lehmann-Waffenschmidt, M. (2022) Evolution und Kontingenz – von der komparativen Evolutorik zur Kausalitätsanalyse, in E. Ökonomik, K. Wegbereiter and S. Anwendungsfelder (eds.), Hrsg. M. Lehmann-Waffenschmidt und M. Peneder. Springer Verlag, pp. 51–66. Ortmann, G. (1995) Formen der Produktion, Westdeutscher Verlag. Witt, U. (2004) Persistence and Change - Is Economic Evolution Theoretical? Erwägen-WissenEthik, 15(1).

184

14 THE FIRM AS AN EXPERIMENTAL DECISION MAKER Gunnar Eliasson

14.1 The rational foundation of the experimentally organized firm Business firms organize their decision systems in relation to the economic environment they operate in. Their competence capital therefore relates directly to their potential performance in that environment. Hence, economists who theorize about, or model firm behavior should be clear about in which “theoretical” market environment their firm is supposed to operate. For the economic theorist, this environment has long been that of the Walrasian (static) general equilibrium (GE) model, in which neither dynamic markets nor meaningfully characterized firms can exist, and if they do they are as a rule identical (“representative”), atomic in size, rest comatose in static equilibrium, cannot individually affect each other’s business, and are in general without interest. One reason for economists to be interested in such firms is their preoccupation with the micro foundations of received macroeconomic models, an ambition which by its very nature is liable to lead nowhere (Weintraub 1979). Coase (1937) placed his business agent, his “firm”, in a static general equilibrium environment, and was able to work out the principles that marked out its boundaries to the market. The dynamic problem of how those boundaries change as market competition changes, he left unanswered. Introducing the notion of an experimentally organized market economy and a financially defined business entity, run top down as a competent team, I will add a dynamic, or rather experimental dimension to the theory of the firm. (Eliasson 1987, 1990, 1991a, 2021). Markets are populated by incumbent firms and new firms that enter markets. They compete with each other for customer attention, for human capital or other inputs, and for financial resources, and in so doing, they integrate within their business plans what they know about markets, including what they know about all other firms doing the same. All firms therefore discipline each other. Multidimensional competition based on individual firms’ entrepreneurial capacity and expectations of each other’s behavior, I propose, will never result in the external market clearing solution of the Walras – Arrow – Debreu static WAD model (Eliasson 1992), because such a solution doesn’t exist in a model with heterogeneous and non-atomistic firms, with autonomous price setting leverage and guided by different business plans, the

DOI: 10.4324/9780429398971-16

185

Gunnar Eliasson

behavior of which affects the market circumstances of other firms. One important (transactions) cost of economic development under such a market regime is exit forced by competition (Eliasson & Eliasson 2005) resulting in endogenous structural change. Ex post ex ante differences in price outcomes will affect firms’ price and quantity decisions. Hence, firm behavior should preferably be modeled as experimentally organized as I observed in my field study (1976a) on business economic planning and decision making. I have therefore outlined an alternative evolutionary model of what I call an Experimentally Organized Economy (EOE, Eliasson 1976b, 1977, 1987, 1991a, 2017b, 2018 and a parallel 2021 paper) that defines the market habitat of firms. In the EOE the appropriate mode of behavior of firms, or business organizations, is as experimentally organized decision teams that constantly design, enact and test business experiments in markets (Eliasson 1990, 1996:Chapter III). Such teams are vested in a financially defined organization with vague and constantly changing interfaces with markets. They cumulate earnings and share the wealth created through agreed upon property rights contracts. Coase (1937) modeled his firm’s external interface with the markets to conform to the principles of the static general equilibrium model. My experimentally organized firm will feel better at home in an EOE.

14.2 The nature of entrepreneurial competition The properties of the economy wide model depend on how micro agents relate to one another in markets, how their market perceptions are integrated into their business plans, and how economy wide market reactions feed back on agents. In an EOE there are however no stable interfaces between the firm and the markets. In firms I have studied, organizational change, and therefore also the outer boundaries of the firm, are part of strategic decisions. Mergers and acquisitions (M&As) and divestments are vehicles to achieve longterm business objectives (Eliasson & Eliasson 2005). Dynamically the firm is therefore a meso entity using Dopfer’s (2012) terminology. The distinguishing feature of an EOE is its vast economic universe that expands from being explored See (Eliasson 1987, 2018, 2021) populated by a large number of competing agents, each largely ignorant of how all others perceive of each other. A second point is that each agent to survive constantly must outdo other agents in innovating its products by accessing superior human capital and financial resources. It does that by exploring the economic universe, discovering new opportunities, and learning from competitors, constantly performing new business experiments to be tested in markets. Incumbent business agents struggle to stay alive as new entrepreneurial entrants subject them to new market challenges. The general unpredictability of such competitive market life makes agents constantly commit business mistakes. New entrants are on average less productive than incumbents. Their performance spread, on the other hand, is much wider. Hence, new entrepreneurial entrants are typically optimistic about their capacities. Surviving entrants, in turn, must be more productive than incumbents and tend to force inferior agents to exit, meaning that the evolution of a growing economy occurs through the turnover of agents (entry and exit), which leaves an endogenously (and irreversibly) changing population of firms in its wake. When the positive flow dominates, we have simulated on a micro based macro model, the economy is growing. Beyond a limit, however, the rate of structural change begins to increasingly disturb the capacity of market prices to selfcoordinate the economy, and the rate of growth comes down. An optimal rate of structural change exits that is compatible with maximum 186

The firm as an experimental decision maker

sustainable economic growth. (Eliasson 1983, 2021, 2022; Eliasson, Johansson & Taymaz 2005). That evolution is not always positive, being often misdirected in corrupt societies, as pointed out by Baumol (1990). Competitive entry and challenges by incumbent firms are what keep the restless competition process going. The question is what directs it. To this I will return in section 4 on competence bloc theory. As I have elaborated in parallel papers (Eliasson 2017b, 2021) business agents in an EOE are in general ignorant of circumstances that may threaten their very existence but are keenly aware of their immediate exposure to both incumbents and entrepreneurial entrants that may outperform them, and therefore plan to leapfrog them through innovation from inferior positions. As I observed above, a competition game so structured has in general no external equilibrium outcome but is constantly nudging its players ahead or out (Eliasson 1996:37ff).1 Two strategies therefore dominate firm behavior in an experimentally organized economy; Strategic long-term radical ambitions to overcome competition through innovative leap frogs, and short-term operational performance to counter predictable challenges as they come. In a viable EOE that is the natural business situation for agents, compelling them to constantly act innovatively and pre-emptively. This is also how endogenous economic growth occurs in my micro to macro model MOSES presented in the companion paper Eliasson (2021). Dynamically competitive markets reinforce such pre-emptive behavior. You don’t have to call in Elon Musk (Tesla) and Jeff Bezos (Amazon) to explain the principles of an EOE. In the EOE normal competition is subtle, but still deadly, and thrives on fear with all agents of being done away with by competitors. What motivates such exceptional entrepreneurship is a separate question. But in the absence of natural and artificial barriers, free competitive entry is a both necessary and sufficient condition for the existence of the evolutionary characteristics of an EOE, which also turn business agents into experimental decision makers to survive. The important policy issue is how to design the institutions, a public good (see below on competence bloc theory) that direct competition towards (socially) productive ends.

14.3

The nature of business competence capital

My study on business planning practices (Eliasson 1976a) found firm management focussed on climbing perceived (ex ante) profit hills, in large firms always conducted through internal budget negotiations where central and financially responsible management forced lowerlevel operations managements, that knows better how to do it, to commit themselves to achieve agreed upon targets. Such negotiations would start with a top-down proposal based on previous profitability performance with an added requirement to improve. I called this principle Maintain or Improve Profit (MIP) targets. Even under normal circumstances and always under conditions of radical market disruption, the perceived profit hills change from all the profit hill climbing going on and hence cause firms to fail to meet targets.2 Competence is the dominant production capital in industry, being largely tacit and embodied in people. This tacit competence capital is best presented in Table 14.1. I therefore begin by discussing two directly observable characteristics of the firm as a (1) competence driven, and (2) experimentally organized decision maker (Eliasson 1990). The notion of an extremely large and varied universe of economic opportunities that expands from being explored, reduces the understanding of the whole of individual agents. All market agents are therefore more or less ignorant of circumstances that may imperil their very existence. A complex environment will make all agents “boundedly rational” in the sense of not being fully informed of what is needed for a controlled outcome. Limited 187

Gunnar Eliasson Table 14.1 Competence Specification of the Experimentally Organized Firm Orientation 1 Sense of direction (business orientation/intuition) 2 Risk willingness Selection 3 Efficient identification of mistakes 4 Effective correction of mistakes Operations 5 Efficient coordination 6 Efficient learning feed back to (1) Source: Eliasson (1990).

understanding is compatible with Herbert Simon’s original definition (1955) of bounded rationality, that individuals and firms make rational, perhaps even profit maximizing decisions, conditioned by what they in fact understand. In my interpretation of the concept firms are not only marginally uninformed, as they may be in some asymmetric information applications of received neoclassical theory, but often, or perhaps even normally, fundamentally ignorant of circumstances that may threaten their very survival, and that even in principle, in an EOE, it cannot be otherwise (see philosophical discussion of the size of the economic opportunities space in my companion paper 2021). Decision makers (Eliasson 1992, 2017a) therefore constantly form mental models of the economic environment they operate in to be capable of coming up with single-valued decisions (or design business experiments) that are constantly updated as they are tested in markets or subjected to new experience. An individual who cannot form such single-valued equilibrium models becomes mentally disturbed. A business organization that fails to pass single-valued instructions down its hierarchy gets chaotic. Extreme heterogeneity both on the sending and receiving (“receiver competence”) ends are thus necessary and sufficient conditions for the existence of “tacitness” in the sense of limited communicability (Eliasson 1986:95, 1990). Operating in a vast opportunities space (previous section), furthermore, means that some agents will always be knowledgeable about matters that most won’t. Hence, in many decision situations, individuals will act on insights that are not only unknown to outsiders, perhaps everybody else, but will also be unable to communicate those insights because of inferior “receiver competence”. This is sufficient to introduce intuition simply as unique insight that cannot be communicated to others. Such business intuition (Item 1 in Table 14.1) gives mental directions to the decision maker about what to do. S/he is however incapable of communicating in full what is on her/ his mind to other participants in the decision process. Having the authority also to decide as an owner or elected executive s/he will appear as excessively risk prone (item 2) to outsiders. This is, however, because of differences in understanding. In reality, s/he is not, because s/he intuitively understands what can be done (Eliasson 2003). Decision makers’ intuition may of course be wrong. Once a decision has been taken it becomes important to identify if a mistake might have been committed, and if so with no delay effectively correct it (Items 3 and 4). Finally, if all checks have tested positive routine management clicks in under Item 5, and whatever has been learned feeds back as experience to enhance the intuitive capacity under 1 (“learning”). Normally, the competencies to orient the business, check for mistakes (selection), and to run it (operations) do not reside in one person. To organize all these capacities 188

The firm as an experimental decision maker

effectively together a team has to be formed (Eliasson 1990). The study of the composition of that team therefore is the way to understand the nature of the competence capital of a business organization. This in turn means that staffing the top competent teams of firms through recruiting and firing of people is the by far most critical task in the business world (Eliasson 1996:35ff). There is a wealth of illustrative cases where this decision process has gone wrong, notably under Items 3 and 4, when business mistakes have been identified late, and even more so, soon identified, but slow to be effectively corrected. Simulations on the agent-based MOSES model in my accompanying paper (2021) provide an eye-opener on the costs and benefits involved in correcting business mistakes and tell something about what kind of business competence is needed. Eliasson & Lindberg (1981) found that even large investment mistakes caused relatively minor social and private costs as long as they were rapidly identified and closed down. The large social and private costs were incurred through carrying on operations in a failed project and depriving society of the services of skilled and competent labor that should have been more productively employed elsewhere.3 Hence, if you believe you have a good business intuition (Item 1 in Table 14.1) dare to be bold (Item 2) and brutal if found out wrong (Items 3 and 4). Then you can engage in many businesses with confidence, fail now and then, learn from your mistakes and still be a winner. And you will be a socially productive market agent.

14.3.1

Extreme heterogeneity means allocation matters

An economy populated by fundamentally but differently ignorant actors means heterogeneity of capabilities, which in turn means that the competence capital determining Item 1 (in Table 14.2) is essentially non quantifiable and asymmetrically distributed. This diversity of faculties of individuals and organizations of individuals means that whatever “section” of that competence spectrum is currently employed on a task, a large part of the total remains redundant (Item 2). Since it is still there, ready to be employed, it can be redeployed on new tasks, hence defining a flexibility in use (Item 3).4 Difficulties of communication because of lack of a similar diversity at the receiving end (Item 4, “tacitness”) makes it impossible to realize the full potential of that flexibility. Market experiments will be needed. The higher up in a decision hierarchy the more critical for the innovative capacity of the entire organization that its executive people have a broad-based receiver competence capable of managing the innovative integration of the increasingly specialized capabilities dominating the hierarchy the further down you get. Hence, the value of the competence capital of an individual or an organization of individuals depends on being understood and identified, and hence on its allocation (Item 5), but is always relative to competing bearers of similar competencies (Item 6). Head hunters, recruitment agencies, HR departments of firms, and Table 14.2 General Characteristics of Competence Capital Competence embodied in individuals or teams of individuals is 1 Extremely heterogeneous and asymmetrically distributed, thus 2 Redundant in any allocation, but also 3 Potentially flexible in its use, and 4 Embodied (tacit) and impossible to communicate in full. As a consequence 5 Its value depends on being understood (receiver competence), how it is allocated, and is 6 Relative to competing contributors of competence Source: Eliasson (1994a).

189

Gunnar Eliasson

labor market agencies all work on unraveling the job potential of individuals to help them find meaningful jobs that match their potential best. Because of the extreme heterogeneity of the competence capital of individuals or teams of individuals (Item 1, Table 14.2) analytical methods to identify its value constantly fail, especially when it comes to unique talent (Eliasson 1994b). The best vehicle for allocating potential talent therefore is a broad-based market for competence with customers exhibiting diverse competencies, and actively searching job seekers. Trial and error in both hierarchies and markets rule the game. People therefore best find their way in the market for competence on their own and learn by moving around (Eliasson 2020).

14.4

Competence bloc theory – The birth, the life, and the death of businesses

The outer limits of a business hierarchy, or a firm, Coase (1937) observed are set where production flows are most efficiently coordinated, within the hierarchy or distributed in the market. A modern firm is however much more than a manufacturing plant (a factory, a “production function”), as is the typical business unit in neoclassical microeconomic theory. Business literature also offers a rich menu of firm concepts and interfaces with particular markets, but they regularly lack both the generality and the links needed to aggregate up via dynamic markets to economy wide levels (for a survey, see Eliasson 1996:III.17). With that done the modern business entity becomes a far more complex (and unstable) composition of activities needed to profitably create and manufacture new innovative products (and services) and take them to final markets. To succeed in doing that over the longer term, during its life a firm has to change its internal composition of activities several times over. The competence bloc in Table 14.3 (Eliasson & Eliasson 1996, 2005, 2009) lists the minimum of such actors with specialized competencies needed, each of which can sometimes be found operating autonomously in a market, or within a corporate hierarchy. Hence, the boundaries of each such activity can be financially defined and are marked out where transactions costs of doing business within a hierarchy, as Coase (1937) proposed, are below those incurred in the market. (The point made is that the competence bloc can be, but rarely is, organized as a centralized hierarchy, but as a typical meso activity in the sense of Dopfer (2012). But since transactions costs depend on technological change, and on the competitiveness of local markets, that also Table 14.3 Actors in the Competence Bloc Who governs the efficiency of selection and scale up Competent customers 1 Competent customers Technology supply in markets for innovation 2 Innovators who develop new technologies and integrate them in new ways Commercializers (scale up) 3 Entrepreneurs, who identify profitable innovations 4 Industrially competent venture capitalists, who recognize and finance promising entrepreneurs 5 Private equity actors, who facilitate ownership change, and exits for venture capitalists 6 Industrialists who take successful entrepreneurial ventures on to industrial scale production (operations management/coordination) Scale down 7 Rational and efficient liquidators (market exit) Source: Eliasson & Eliasson 1996, 2005.

190

The firm as an experimental decision maker

changes, the hierarchical boundaries are unstable. Hence, competence bloc theory offers a blue print of how these organizational dynamics occurs.)

14.4.1

Competent customers make a difference

The customer (Item 1 in Table 14.3) resides at the top of the competence bloc, and all other actors have to relate upwards to the preferences of customers. This also means that the competence of customers to enjoy innovative new and sophisticated products and the capacity of customers to pay for their development set the stage for industrial advance. Burenstam-Linder (1961) found customer sophistication and wealth among industrial economies to be a determinant of national comparative advantage. When producers of military equipment, for instance, are allowed to act freely without political interference to develop sophisticated new weaponry such projects often become platforms for technological spillovers. Customer competence then translates directly into observed technical change (Eliasson 2011, 2017b). When demand and supply curves become dependent on one another the very foundation of the received neoclassical equilibrium model disappears.

14.4.2

Commercialization and scale up

Innovators (Item 2) develop new technologies and combine them with existing ones into new products and production processes and supply them in the markets for innovation. On the demand side entrepreneurs (Item 3) understand the commercial value of innovative new technologies offered and select promising ones for further development.5 Entrepreneurs however often lack the financial resources for that and must team up with industrially competent venture capitalists (Item 4) to continue the commercialization process of tailoring innovative new products such that their commercial value is branded and made understandable for standardized trading in the private equity market (Item 5). The industrial competence of venture capitalists is critical for minimizing the incidence of premature termination of long-term winners (Type II economic mistakes. See below). The private equity operator begins the scale up process to industrial level production and distribution and either prepares an IPO or a sale to a large industrial operator (Item 6). The picture is not complete until the final decline or death (exit) of hierarchies has been covered (Item 7). Recent history of corporate governance is illustrative of how the relative change of transactions costs has changed the organizational forms of scale up. Here the markets for strategic acquisitions and divestures operate as a dynamic restructurer at all levels of the competence bloc (Eliasson & Eliasson 2005). The new millennium furthermore appears to have meant a radical improvement, in the United States in particular, in the supplies of industrially competent finance covering scale up all the way from establishment to industrial scale production (Items 3 through 6), witness the sudden and disruptive emergence of many tech giants such as Alphabet, Amazon, and Tesla. To be noted is that organizational change occurring through the commercialization phase is as much (1) acts of innovation and entrepreneurship as is the original innovation under Item 1, and the bulk of resource use is incurred under scale up. Multifaced receiver competence that minimizes losses of winners during scale up is the critical factor behind holding down development costs both at the firm and the economy wide level.

191

Gunnar Eliasson

14.4.3

Receiver competence

Competence bloc theory (in Table 14.3) lists the minimum of actors with competence needed to identify, select and bring winners out of a wide supply of technologies to industrial scale production and distribution. Together they define the receiver competence of the commercialization markets (Eliasson 1986, 1990). When the competence bloc is vertically complete and horizontally sufficiently varied to minimize the economic consequences of making two types of economic mistakes (Eliasson & Eliasson 1996): keeping losing projects on the books for too long (Mistake Type I) and losing the winners (Mistake Type II), it has reached critical size. Then the competence bloc functions both as a spillover source and an attractor of outside technologies. A promising innovator/entrepreneur can then confidently continue to look for resources and will eventually capture them. Under such positive circumstances, we will observe an increased flow of winners followed by a smaller flow of mistaken economic projects. Economic growth will increase. Many of the selection services of the competence bloc under Items 3 and 4 and the scale up services of Items 5 and 6 in Table 14.3 are internalized within large firms. The different competence contributions are then sequentially coordinated by management within a hierarchy, rather than by competition in markets, which, given the circumstances of the insights required, are likely to lead to inferior matches compared to those of a more broad-based market. On the other hand, if both innovative capacity, management competence, and (financial) resources exist within the same hierarchy, commercialization and scale up will be much faster. One might argue that great technological opportunities coupled with optimism and “animal spirits” to use a term coined by Keynes and argued by Akerlof & Shiller (2009) as necessary to guarantee a rich supply of entrepreneurship “from within the economy”. But are animal spirits also a sufficient condition? A more compelling force than profit incentives may be needed for the small innovator/entrepreneur to overcome the natural and regulatory barriers, and maybe even a mental resistance, to dare to enter the market. As is explained in the companion paper Eliasson (2021), innovation spurred by fear of being overtaken by competition is the missing link that makes the evolutionary model of an Experimentally Organized Economy complete.

14.5 Coase revisited – The firm as an experimental decision maker A business hierarchy has a horizontal dimension in which different tasks, such as production or distribution, are coordinated in the short term and a vertical longer-term dynamic dimension in which new technologies or products are created, selected, improved upon, and taken to market. The two tasks are fundamentally different and demand different competencies from management. Competence bloc theory is concerned with both, but specifically addresses the second dynamic dimension. The end outcome of a competence bloc sequence should be a winner that has been taken to industrial scale production and distribution and entered a phase of routine management under Item 6 in Table 14.3. Over time, a business hierarchy changes its composition through acquisitions and divestments and different rates of change of its component activities. Coase (1937) gave a snapshot picture of the firm when he explained its outer limits visa a vis the market in terms of relative transactions costs of coordinating its production internally compared to doing it over the market. The competence bloc explains such compositional change over time. In a dynamic perspective, a firm or a business hierarchy therefore is not a well-defined entity, but a loosely structured aggregate, or a meso entity to use the terminology of Dopfer (2012).

192

The firm as an experimental decision maker

Each function of the competence bloc in Table 14.3 relates to well-defined interfaces with a market. They can in principle be managed as autonomous business entities in competition with others in financial markets, but synergistic integration between some of them sometimes confers productive benefits to the whole. Since the 1990s, for instance, financial markets have gone through a combined technical and organizational outsourcing revolution. Some of the market intermediators of the competence bloc did not exist a few decades ago and still don’t even in advanced market economies, and definitely did not when Coase (1937) wrote his article. Private venture capitalists (VCs) of the professional industrially competent kind that can be found in Silicon Valley in California are rare in continental Europe, where industrial bank finance still dominates, and public venture capital provision is more typical (Rybsczynski 1993, Eliasson 2003).

14.5.1

The competence bloc integrates the short and the longer term

While short-term horizontal coordination for “maximum” immediate profits is one thing, managing the longer term vertically through the competence bloc demands radically different competencies, and getting both the long-term and the short-term right is a highly complex management task (Eliasson 2005). As is well known from industrial practice, the two sides often conflict when managed under the same central hierarchy. The efficient static production organization of a large manufacturing business often kills internal incentives for innovation and entrepreneurial initiatives. An advanced manufacturing firm set on a radical strategic program to enter a new complex technology rarely has all its technological know-how in-house. The market for strategic acquisitions has therefore become an important vehicle for corporate organizational change and the transformation of hierarchies by supplying complementary technologies to large firms lacking internal innovative capacities to reorganize themselves for new products and markets.

14.5.2

The firm as a competent team and experimental decision maker

When widely different competencies are demanded, separation of responsibilities through reorganization is often needed to manage operations and innovation within the same hierarchy. To make two widely different and competing teams of different people and with different responsibilities coordinate their ambitions is however difficult in the unpredictable market environments of experimentally organized industrial economies. Since the ambition of top CHQ central management often is to control both within the same firm hierarchy, failure on both counts is common. Whether control-minded central CHQ management likes it or not, and however large resources are invested in analysis and internal argument, in the end even the decisions in very large global and diversified business organizations become characterized as business experiments to be tested in markets, and hence failure prone. The reasons for failure can often accurately be sorted out ex post (through analysis) as inattention to something that should perhaps have been thought of to begin with. The reasons for business success are more complicated to sort out ex post because they often rest on a unique “entrepreneurial insight” that an individual or a team of individuals has had that cannot logically be derived from past experience and that the individual/team has both dared and been capable to realize as a business experiment. 193

Gunnar Eliasson

My original notion of the firm as a competent team and an experimentally organized decision maker therefore makes the markets for specialized executive competencies the critical factor in the study of industrial development (Eliasson 1991b). Global recruiting of specialized talent has become a necessary source of talent for advanced firms. In fact, to put a value on such a firm, its critical capital being invisible and locationally mobile human embodied competence becomes synonymous with evaluating the people in charge, and to reorganize a firm becomes synonymous with reorganizing its people with competence. This is of course understood by practical managers who therefore normally do it through experimental hiring and firing. At the top competent team which dominates downstream allocations of competence, it often means the same as reshuffling of ownership, which to succeed smoothly requires an active and industrially competent market in ownership contracts (Eliasson & Eliasson, 2005). The allocation of the dominant and most important production capital within hierarchies and over the market is therefore best left to the experimental trial and error intermediation of the many agents in the markets for competence

Notes 1 Also of theoretical interest should be to note that in the model of an EOE of my companion 2021 paper, such an equilibrium was demonstrated not to exist. 2 The MIP targeting principle was directly coded into the firm models of the MOSES evolutionary economy wide model that is the main subject matter in my companion paper Eliasson (2021). Also see Eliasson (1977, 1978:185–192, 1980). From the point of view of economic doctrines, it might be of interest to know that under neoclassical fixed structure assumptions, MIP targeting would approximate profit or rather wealth maximization behavior. In the MOSES model of an EOE, the constant failure of agents to realize expectations and plans – a Stockholm School feature – becomes a normal (transactions) cost of economic growth. 3 Huge social costs were incurred in the form of lost output when government subsidized the (in practice) bankrupted Swedish shipyards for many years in the wake of the oil crises of the 1970s. By failing to shut down the yards the economy at large was deprived of thousands of highly skilled workers. The social costs incurred were calculated on the MOSES model ( Carlsson 1983, Carlsson et al. 2018). 4 Even though there are of course transactions costs associated with exercising that flexibility. 5 The entrepreneur possesses a unique commercial understanding of the innovator’s technological idea, an understanding that will be largely tacit, but perhaps be shared by more than one actor. Often the innovator and the entrepreneur are the same.

References Akerlof, George A., & Robert J. Shiller, 2009. Animal Spirits – How Human Psychology Drives the Economy, and What Matters for Global Capitalism. Princeton, Oxford: Princeton University Press. Baumol, William J., 1990. “Entrepreneurship: Productive, Unproductive and Destructive,” JPE, 98(5): 893–921. Burenstam-Linder, S., 1961. An Essay on Trade and Transformation. Uppsala: Almqvist & Wiksell. Carlsson, Bo, 1983. “Industrial Subsidies in Sweden: Macro-Economic Effects and an International Comparison,” Journal of Industrial Economics, XXXII (1) (Sept.): 9–14. Carlsson, Bo, Gunnar Eliasson, & Karolin Sjöö, 2018. “The Swedish Industrial Support Program of the 1970s Revisited,” Journal of Evolutionary Economics, 28(4), (Sept): 805–835. Coase, R H, 1937. “The Nature of the Firm,” Economica, New Series, IV (13–16) (Nov.): 386–405. Dopfer, Kurt, 2012. “The Origin of Meso Economics – Schumpeter’s Legacy and Beyond,” Journal of Evolutionary Economics, 22(1) (January): 133–160.

194

The firm as an experimental decision maker Eliasson, Gunnar, 1976a. Business Economic Planning - Theory, Practice and Comparison. London: John Wiley & Sons. Eliasson, Gunnar, 1976b. “A Micro-to-Macro Model of the Swedish Economy, Preliminary Documentation,” (with the assistance of Gösta Olavi and Mats Heiman). Economic Research Report B15. Stockholm: Federation of Swedish Industries. Eliasson, Gunnar, 1977. “Competition and Market Processes in a Simulation Model of the Swedish Economy,” American Economic Review, 67(1): 277–281. Eliasson, Gunnar, 1978, (ed). “Micro-to-Macro Model of the Swedish Economy,” Conference Reports, 1978:1. Stockholm: IUI. Eliasson, Gunnar, 1980. “Experiments with Fiscal Policy Parameters on a Micro-to-Macro Model of the Swedish Economy,” In R.H. Haveman & K. Hollenbeck (eds.), Microeconomic Simulation Models for Public Policy Analysis, Vols. 1 & 2. New York, London: Academic Press. Eliasson, Gunnar, 1983. “On the Optimal Rate of Structural Adjustment,” In G. Eliasson, M. Sharefkin & B.-C. Ysander (eds.), Policy Making in a Disorderly World Economy, Conference Reports, 1983:1. Stockholm: IUI. Eliasson, Gunnar et al, 1986. Kunskap, information och tjänster – en studie av svenska industriföretag (Knowledge, Information and Service Production – A Study of Swedish Manufacturing Firms). Stockholm: IUI. Eliasson, Gunnar, 1987. Technological Competition and Trade in the Experimentally Organized Economy. Research Report No. 32. Stockholm: IUI. Eliasson, Gunnar, 1990. “The Firm as a Competent Team,” Journal of Economic Behavior and Organization, 13(3) (June): 275–298. Eliasson, Gunnar, 1991a. “Modeling the Experimentally Organized Economy - Complex Dynamics in an Empirical Micro-Macro Model of Endogenous Economic Growth,” Journal of Economic Behavior and Organization, 16(1–2): 153–182. Eliasson, Gunnar, 1991b. “Financial Institutions in a European Market for Executive Competence,” Chapter 7 in C. Wihlborg, M. Fratianni, & T.A. Willet (eds.), Financial Regulation and Monetary Arrangements after 1992. Amsterdam: Elsevier Science Publishers B.V. Eliasson, Gunnar, 1992. “Business Competence, Organizational Learning and Economic Growth Establishing the Smith-Schumpeter-Wicksell Connection,” In F.M. Scherer, & M. Perlman (eds.), Entrepreneurship, Technological Innovation, and Economic Growth: Studies in the Schumpeterian Tradition. Ann Arbor: University of Michigan Press:251–277 (First presented at the 1990 Joseph A. Schumpeter Society meeting in Airlie House, Virginia, USA, June 3–5, 1990). Eliasson, Gunnar, 1994a. “Educational Efficiency and the Markets for Competence,” European Journal of Vocational Training, (2), 1994:5–11 Eliasson, Gunnar, 1994b. Markets for Learning and Educational Services - A Micro Explanation of the Rôle of Education and Competence in Macroeconomic Growth. Paris: OECD, DEELSA/ED/CERI/ CD, (94) 9 (04-Nov.). Eliasson, Gunnar, 1996. Firm Objectives, Controls and Organization - the Use of Information and the Transfer of Knowledge within the Firm. Boston/ Dordrecht/London: Kluwer Academic Publishers. Eliasson, Gunnar, 2003. “The Venture Capitalist as a Competent Outsider,” In Kari Alho, Jukka Lassila, & Pekkaa Ylä-Anttila, (eds.), Economic Research and Decision Making. Helsinki: ETLATaloustieto oy. Eliasson, Gunnar, 2005. “The Nature of Economic Change and Management in a New Knowledge Based Information Economy,” Information Economics and Policy, 17: 428–456. Eliasson, Gunnar, 2011. “Advanced Purchasing, Spillovers and Innovative Discovery,” Journal of Evolutionary Economics, 21.1–4:121–139. Eliasson, Gunnar, 2017a. “Micro to Macro Evolutionary Modeling – On the Economics of Self Organization of Dynamic Markets by Ignorant Actors,” In U. Cantner, & A. Pyka (eds), Foundations of Economic Change – Behavior, Interaction and Aggregate Outcomes, ECONOMIC COMPLEXITY AND EVOLUTION. Berlin, Heidelberg, New York: Springer Eliasson, Gunnar, 2017b. Visible Costs, Invisible Benefits. New York, Dordrecht, Heidelberg, London: Springer. Eliasson, Gunnar, 2018. “Why Complex, Data Demanding and Difficult to Estimate Agent Based Models?- Lessons from a Decades Long Research Program,” International Journal of Microsimulation, (2018)11(1): 4–60.

195

Gunnar Eliasson Eliasson, Gunnar, 2020. Sverige och den Globala Marknaden för Kompetens (The Swedish Economy and the Global Market for Competence). Stockholm: Ekerlids Förlag. Eliasson, Gunnar, 2021. “The role of markets in economic development” manuscript prepared for Kurt Dopfer (ed)”, Evolutionary Economics, forthcoming Elgar. Eliasson, Gunnar, 2022. Sustainable Finance and Disruptive Scale-up”, paper prepared for the Landmark Conference on Sustainable Finance in memory of professor Clas Wihlborg, Marstrand, June 13–14, 2022. Eliasson, Gunnar, & Åsa Eliasson, 1996. “The Biotechnological Competence Bloc,” Revue d’Economie Industrielle, 78–40. Trimestre: 7–26. Eliasson, Gunnar, & Thomas Lindberg, 1981. “Allocation and Growth Effects of Corporate Income Taxes,” In G. Eliasson, & J. Södersten (eds.), Business Taxation, Finance and Firm Behavior, Conference Reports 1981:1. Stockholm: IUI: 381–435. Eliasson, Gunnar, & Åsa Eliasson, 2005. “The Theory of the Firm and the Markets for Strategic Acquisitions,” In U. Cantner, E. Dinopoulos, & R.F. Lanzilotti (eds.), Entrepreneurship. The New Economy and Public Policy. Berlin, Heidelberg, New York: Springer: 91–115. Eliasson, Gunnar, Dan Johansson, & Erol Taymaz 2005. “Firm Turnover and the Rate of Macroeconomic Growth”, Chapter VI In G. Eliasson (ed.), The Birth, the Life and the Death of Firms‐The Role of Entrepreneurship, Creative Destruction and Conservative Institutions in a Growing and Experimentally Organized Economy. Stockholm: The Ratio Institute: 305–356. Simon, Herbert A., 1955, “A Behavioral Model of Rational Choice,” Quarterly Journal of Economics, 69: 99–118. Rybsczynski, T.M., 1993. “Innovative Activity and Venture Financing: Access to Markets and Opportunities in Japan, the US and Europe,” In Richard H. Day, G. Eliasson, & C. Wihlborg (eds.), The Markets for Innovation, Ownership and Control. Stockholm, Amsterdam, North Holland: IUI. Weintraub, E. Roy, 1979. Microfoundations- The Compatibility of Microeconomics and Macroeconomics. Cambridge: Cambridge University Press.

196

15 EVOLUTIONARY ECONOMICS, ROUTINES, AND DYNAMIC CAPABILITIES David J. Teece

15.1

Introduction

The perspective of evolutionary economics has had a decided impact on strategic management, primarily through the resource-based view of the firm and the dynamic capabilities framework (Helfat, 2018). In this essay, I will consider the relationship between dynamic capabilities and evolutionary economics. Evolutionary economics intersects most poignantly with the capabilities literature around the concept of routines. Winter (2000, p. 983) defines an organizational capability as: a high-level routine (or collection of routines) that, together with its implementing input flows, confers upon an organization’s management a set of decision options for producing significant outputs of a particular type. The presence of “outputs” in this definition make clear that it applies to operational (or “ordinary”) capabilities. Management is at least implicitly involved here: coordinating routines, “implementing input flows,” and choosing the type and level of outputs. It appears to be the routines that define the scope of managerial activity in evolutionary economics. Another type or level of capability is required to account for how firms add entirely new ordinary capabilities and eliminate old ones. Teece, Pisano, and Shuen (1990, 1997) called this the “dynamic capabilities approach.” Although Nelson has not written much about dynamic capabilities, he was quick to see their potential, noting in response to the initial 1990 working paper version of Teece, Pisano, and Shuen (1997) that: the ‘dynamic capabilities’ view of firms being developed by scholars in the strategy field can be seen to be important not only as a guide to management, but also as the basis for a serious theory of the firm in economics. It, when embedded in an evolutionary theory of economic change, instructs us regarding ‘Why do Firms Differ, and How Does it Matter?’ (Nelson, 1991, p. 72)

DOI: 10.4324/9780429398971-17

197

David J. Teece

Winter, by comparison, has written a fair amount about dynamic capabilities, but almost always with a view to constraining them within the bounds of routines. When he first examined dynamic capabilities in depth, he simply applied his earlier definition of all capabilities as routines, i.e., “behavior that is learned, highly patterned, repetitious” (Winter, 2003, p. 991). When revisiting the topic later with Helfat, Winter endorsed a more expansive definition of dynamic capabilities as the capacity of a firm to “alter how it currently makes its living” (Helfat and Winter, 2011, p. 1244). However, from a strategic management perspective (and, ultimately, as a matter of economics), the routines-based approach remains too restrictive to the extent that it precludes a role for entrepreneurial decision making. Winter (2003) was presumably afraid that doing otherwise would require him to embrace “ad hoc problem solving.” Perhaps that is also the reason Helfat and Winter (2011, p. 1248) touch on the “dynamic capabilities of top managers” only in passing, despite Helfat’s pioneering work on the topic (Adner and Helfat, 2003). Yet, as I will make clear in what follows, strategic decisions of dynamically capable managers are vital to the fates of firms and the dynamics of industries. As the contrast between Nelson’s and Winter’s responses to dynamic capabilities makes clear, evolutionary economics is not uniform in how it views capabilities and the prospects for strategic renewal. There is a fundamental tension in the evolutionary approach over the relative importance of continuity versus novelty, as represented by the separate outputs of its two chief pioneers. Whereas Winter’s view emphasizes routines and the cost of changing them (e.g., Winter, 2003), Nelson has been more focused on the ability of firms to innovate in technology and organization and to create competitive advantage (e.g., Nelson, 1991). Since Nelson’s early assessment of dynamic capabilities as a potential element within an evolutionary model, I have been constructing an evolution-consistent framework of the firm built around dynamic capabilities. I believe the dynamic capabilities framework can serve as a vehicle for clarifying the relationship between routines and innovation. Combining them in a framework in which entrepreneurship and strategy also have their place can better account for how firms change—and how they change industries. The essay proceeds as follows. I start by recapping the separation between routines that evolve and one-off managerial decisions. This is followed by sections which argue that entrepreneurship is hard to capture in an evolutionary framework, provide examples of it in its pure form, and describe its role in the dynamic capabilities framework as part of “entrepreneurial management.” Next, I contrast dynamic capabilities and the evolutionary model of the firm in more detail, particularly with respect to innovation, change, and strategy. A brief section concludes.

15.2

Can routines do it all?

In Nelson and Winter’s (1982) pathbreaking evolutionary model of the firm, the primary activities are building and exploiting knowledge assets, using organizational “routines.” As noted, organizational learning, in the form of processing information and solving problems, leads to new knowledge and improved routines. As routines evolve, each firm generates a unique trajectory. Some firms are better at learning, and some firms learn the wrong lessons and eventually fail. Despite several decades of research on routines, much of the literature remains enigmatic (see Becker, Lazaric, Nelson, and Winter, 2005). Some observers find contradictions, 198

Evolutionary economics, routines, and dynamic capabilities

claiming that the same “recipes” support different ways of performing activities (e.g., Kenney and Florida, 1991; Adler, 1993). I do not doubt the power and importance of routines at the operating level. Organizations need routines to get the day-to-day work done. They economize on bounded rationality. The dynamic capabilities framework recognizes routines as the very essence of operational (ordinary) capabilities. In general, though, the concept of routines is simply not rich enough to account for the most important features of the firm. The inherent nature of the routine-based evolutionary approach is that search tends to be “close in” or local. Routines improve through repetition and through mutation/perturbation plus internal selection. Business can also benchmark their routines/processes, identifying and adopting industry “best practices.” Rarely, however, do all firms end up using the same routines/practices—or practicing them with equal skill (Bloom, Genakos, Sadun, and Van Reenen, 2012). Considerable dispersion occurs, with only a few “frontier firms” that develop and/or copy best-practice routines. In terms of building a model of the economy, the evolutionary approach can therefore point to routine-based capabilities alone as a primary explanation for differences in firm performance. But this excludes multiple major sources of inter-firm heterogeneity. The top management team, or corporate culture and structure for that matter, are very much in the background. Yet any reading of business history of the past 150 years would be incomplete indeed if it did not include the activities of great innovators and integrators such as Henry Ford, Thomas Edison, Charles Merrill, Sam Walton, Walt Disney, Bill Gates, Steve Jobs, and Elon Musk, who each transformed one or more industries by pursuing a singular vision with unusual skill. While these individuals had exceptional talents, every company has the potential to strengthen its dynamic capabilities by diffusing an entrepreneurial state of mind throughout its structure. The entrepreneurial decision making critical to the exercise of dynamic capabilities lies outside the purview of evolutionary economics. While the honing of routines is vital for “doing things right” (i.e., efficiently), routines are just one contributor to “doing the right things” (i.e., figuring out where and how to invest for the future). It is not evident that routines are helping Elon Musk, CEO of Space-X, decide how to get a crew to Mars and make money as the program evolves. Nor is it clear that routines helped Steve Jobs create the iPhone ecosystem; or that they played any role in helping Jeff Bezos at Amazon decide to offer commercial cloud computing services. These decisions about “doing the right things” lie at the core of business strategy. The quality of these decisions, along with luck, shape the evolution/progress of the firm, and they are anything but routine. Although signature routines may be involved (Gratton and Ghoshal, 2005), they still require some amount of entrepreneurial decision making. This in turn requires an entrepreneurial state of mind: a high level of creativity, a high level of energy, relatively low risk aversion, the ability to see new opportunities and new combinations, a sense of urgency and a bias for action. Routines (and capabilities), on the other hand, are equated with a recipe or a program, implying that they could potentially be carried out by simple automatons. Nelson and Winter (1982, p. 94) made clear that they “emphasize the automaticity of skillful behavior and the suppression of choice that this involves.” Yet these tasks must still be coordinated and managed. Nelson and Sampat (2001) call routines the “physical technology” and their deployment the “social technology.” This deployment—or what I call “seizing” in dynamic capabilities terms—is absent or underdeveloped in evolutionary models of the firm. 199

David J. Teece

It is unclear if dynamic capabilities can even be modeled in an evolutionary framework. There are any number of studies taking an evolutionary approach to capabilities, but the capabilities involved are, for the most part, what I call “ordinary capabilities” or, in some cases, what Winter calls “first-order capabilities” (and others call “metaroutines”), i.e., involved in changing existing capabilities (and routines) in incremental ways (Winter, 2003; Adler, Goldoftas, and Levine, 1999; Hoopes and Madsen, 2008). Most ordinary capabilities can be benchmarked to industry best practices. However, the ability to do so indicates the likelihood that the benchmarked capabilities can be bought or imitated by rivals and hence are unlikely to be a source of competitive advantage. Some ordinary capabilities are specialized and proprietary and can serve as a competitive wedge for an extended period. These employ “signature processes” (Gratton and Ghoshal, 2005) that grow out of a firm’s particular history and context and set a new performance standard. Toyota’s lean production model, a tightly integrated set of processes, is one example (Womack, Jones, and Roos, 1990). Important elements of the “Toyota Production System,” such as just-in-time supplies and quality control circles, gradually diffused to rivals; but Toyota’s unique implementation provided the automaker a source of competitive advantage for decades (Fruin and Nishiguchi, 1990) and, arguably, still does so. Teecian dynamic capabilities, by comparison, are higher-order bundles of organizational routines and managerial decisions that drive the strategic activities of the business enterprise competing in regimes of deep uncertainty (Teece, 2012; Teece, Peteraf, and Leih, 2016). I see dynamic capabilities residing not primarily in first-order change routines but in the characteristics of the top management team, in the organizational culture and structure, and in supporting organizational routines which are under the stewardship of top management. As noted earlier, evolutionary theorists are reluctant to adopt this perspective, perhaps for fear it would require one to embrace what Winter (2003) characterizes as “ad hoc problem solving.” For Winter (2003, p. 991), “Brilliant improvisation is not a routine” However, a series of brilliant improvisations may indicate the existence of a quasi-routine for how an entrepreneurially led organization solves problems. Furthermore, what may appear ad hoc to someone else may have a logic of its own not readily understood by others—a notion captured to some degree by Rumelt’s concept of uncertain imitability (Lippman and Rumelt, 1982). Part of the problem may be that the key concepts of evolutionary economics and organizational routines developed in a world before the second digital age and the “big data” revolution. In today’s world, the set of production possibilities, monetization strategies, and competitive opportunities arising from large, diverse, and numerous datasets is vast, which augments complexity and deepens uncertainty. The payoffs to high-level routines for learning and search are relatively modest; but the payoffs to strong dynamic capabilities for learning and search are substantial. The evidence is not hard to find. Silicon Valley firms have been making large investments in AI systems designed to “sense, think, and act” in complex data environments. This is analogous to the learning ethos of dynamic capabilities, namely sensing, seizing, and transforming. Perhaps there is a definitional problem. Winter (2003) divides activities and decisions in a binary way; they are either routine or ad hoc. But entrepreneurial decision making is a hybrid, involving creative decisions that are informed by and that control repeatable and repeated business processes. Arndt and Pierce (2018) note that dynamic capabilities involve a number of activities that fall into this hybrid area, including business model development, long-run investment choices, development of complementary and cospecialized assets, and 200

Evolutionary economics, routines, and dynamic capabilities

asset orchestration. To the uninitiated, decisions in these areas might seem ad hoc because they are one-offs. Maybe the need for repeated “ad hoc” decisions in a given context (e.g., business model development) amounts to a sort of entrepreneurial routine requiring a creative, non-routine state of mind. A description by Steve Jobs of the innovation process at Apple conveys the need to balance routines and the non-routine creative capabilities of the organization: Apple is a very disciplined company, and we have great processes. But that’s not what it’s about. Process makes you more efficient. But innovation comes from people meeting up in the hallways or calling each other at 10:30 at night with a new idea, or because they realized something that shoots holes in how we’ve been thinking about a problem. It’s ad hoc meetings of six people called by someone who thinks he has figured out the coolest new thing ever and who wants to know what other people think of his idea. And it comes from saying no to 1,000 things to make sure we don’t get on the wrong track or try to do too much. We’re always thinking about new markets we could enter, but it’s only by saying no that you can concentrate on the things that are really important. (cited in Burrows, 2004) Jobs seems to say that, while Apple’s ordinary capabilities are based in processes, its product development is a looser set of activities and non-routine cognition. Parts of this process might appear ad hoc, but they collectively amount to an extremely valuable capability that is ultimately governed by top management decisions about which products to pursue.1 Winter himself seems to have come some distance toward accepting that higher-order capabilities encompass not just routines but also managerial decisions. In a 2017 article, he explicitly discussed the distinction between what he calls System 1 (automatic routines) and System 2 (deliberative decisions), adapted from concepts introduced in the psychology research of Daniel Kahneman (2011). In developing his own version with respect to dynamic capabilities, Winter noted that “If, hypothetically, I were to pursue it, I would still be arguing for tuning the System 1 emphasis up a little and System 2 down a little, relative to Teece” (Winter, 2017, p. 73). In this paper, I am of course suggesting that one tune the System 2 emphasis higher—at least in environments of deep uncertainty or VUCA (volatility, uncertainty, complexity, and ambiguity).

15.3

How much room for entrepreneurship?

The managerial decisions behind dynamic capabilities require a distinctive state of mind: that of an entrepreneur. In their 1982 book, Nelson and Winter quote William Baumol to the effect that mainstream economic theory “offers us no promise of being able to deal effectively with the description and analysis of the entrepreneurial function” (Baumol, 1968, p. 68 cited by Nelson and Winter, 1982, p. 32). While their purpose is to argue that the passive maximization of neoclassical economics is a delusion, they are more interested in the bounded rationality of the manager than in the role of the entrepreneur. In fact, entrepreneurs are scarcely mentioned again, apart from a brief discussion of Schumpeter on page 277, where they are acknowledged as the agents within the economic system who drive technical progress. But we are a long way from a description, much less an analysis, of the entrepreneurial function. 201

David J. Teece

Interestingly, in an essay that Winter wrote in the 1960s which was not published until 2006,2 he restates Schumpeter’s view of the entrepreneur as “the leader who leads the firm to new techniques … the carrying out of new combinations.” There is then a direct quote from Schumpeter: “Carrying out a new plan and acting according to a customary one are as different as making a road and walking on it” (Schumpeter, 1934/1949, p. 85). Clearly, Schumpeter saw the entrepreneurial function in the economy as different from the routine one. Yet Winter claims (2006, p. 137) that Schumpeter sees “the essential continuity between these instances of dramatic innovation and the smallest sort of adaptation to changing conditions” (emphasis in the original). The citation he provides, however, shows something quite different. Schumpeter is hypothesizing not about activities like innovation and adaptation but about the distribution of skills. He suggests that entrepreneurial skill, which he likens to singing (everyone can do it at least a little bit, even if only badly), is distributed among people along something like a normal distribution. Schumpeter notes that, even if the exact cutoff points is hard to identify, there is a material difference between the best and the rest. In the words of his analogy: Only in this [top] quarter are we struck in general by the singing ability, and only in the supreme instances can it become the characterising mark of the person. Although practically all men can sing, singing ability does not cease to be a distinguishable characteristic and attribute of a minority. (Schumpeter, 1934/1949, p. 82, fn2) Thus, where Winter sees continuity, Schumpeter sees a difference in kind. Making a road is quite different from walking on one. This is not just a matter of comparing a high-order routine to a simpler one. By emphasizing continuity, Winter has essentially removed the entrepreneur from Schumpeter. His concern, at least in that essay, was in how production methods change, not in how firms prepare for the future and look for the next big thing. In capabilities terms, this is the difference between ordinary and dynamic capabilities.

15.4 What great entrepreneurial management looks like In this regard, it is instructive to examine the approaches/heuristics that animate one of the greatest living entrepreneur-managers, Elon Musk. As reported by blogger Tim Urban (2015), who interviewed Musk extensively, Musk refers to one of his key habits of mind as “reasoning from first principles”: I think generally people’s thinking process is too bound by convention or analogy to prior experiences … . But that’s just a ridiculous way to think. You have to build up the reasoning from the ground up—“from the first principles” is the phrase that’s used in physics. You look at the fundamentals and construct your reasoning from that, and then you see if you have a conclusion that works or doesn’t work, and it may or may not be different from what people have done in the past. According to Urban, after Musk reasons out his goals for a project and a strategy, “[h]e tests … them, continually, and adjusts them regularly based on what he learns.”

202

Evolutionary economics, routines, and dynamic capabilities

Urban contrasts Musk’s way of thinking with how most people deal with the world as “the difference between a cook and a chef.” Cooks practice routines. Chefs innovate: The chef reasons from first principles, and for the chef, the first principles are raw edible ingredients. Those are her puzzle pieces, her building blocks, and she works her way upwards from there, using her experience, her instincts, and her taste buds. The cook works off of some version of what’s already out there—a recipe of some kind, a meal she tried and liked, a dish she watched someone else make. As it happens, Winter (2006) has also used the “cook” metaphor. Initially, he focuses on the need to combine inputs with appropriate human capital. But he also considers innovation: Talented cooks can create new dishes by thinking up a new taste and then writing the recipe; talented composers can hear their music before they write down the required inputs of violin notes and horn notes; talented engineers can develop an idea about the physical basis for a new device into a detailed design. In all of these cases, the talent involved includes a sweeping grasp of the behavior of the relevant materials under a very wide range of conditions. Without actually baking the cake or playing the symphony or building the device, tentative solutions to the problem can be “tried out” and modified or rejected as necessary. There is no simple answer to the question of where this knowledge comes from. Some of it—the engineer’s understanding of physical laws or the composer’s understanding of harmony—is theoretical knowledge. Some of it is the fruit of long personal experience with the general class of problems involved, and some is obtained from reports of the experience of others, or direct contact with their work. (Winter, 2006, p. 133) Winter clearly saw the difference between a chef (or “talented cook”) and a plain cook who only knows how to follow recipes. But his interest is in different levels and types of productive knowledge within the firm, not in how the talented cook can start a new business. Musk has noted that reasoning from first principles isn’t desirable or necessary at all times. Most of the time, “copying what other people do with slight variations” is enough. “Otherwise, mentally, you wouldn’t be able to get through the day” (cited in Popomaronis, 2020). Put differently, routines help to accomplish day-to-day work and economize on bounded rationality. But strategy work requires a different approach. Teece, Peteraf, and Leih (2016) make a case for abductive reasoning, which is close to Musk’s reasoning from first principles, at least as he applies it. Of course, reasoning from first principles, or abductive reasoning, could be thought of as a routine. But this misrepresents the strategy process, which, if written down, would look something like “gather all the facts, have a brilliant insight, work really hard, repeat.” The problem with this is that not every entrepreneur-manager can have insights on demand, although a select few seem to. There are creative and insight-dependent components that cannot be routinized. Entrepreneurship isn’t an adaptation to environmental conditions; it’s the making of “new combinations” (Schumpeter, 1934/1949, p. 66). Musk, Steve Jobs at Apple, Jeff Bezos at Amazon, or Page and Brin at Google weren’t adapting to environmental changes, either at the founding of their companies or as the years passed. They were building futures they had envisioned. 203

David J. Teece

The forging of organizational routines and capabilities may occur after investment decisions are made. Existing routines (or their absence) need not constrain the decision space. Although it might have been safer to partner with an existing auto maker, Musk created his own routines for Tesla’s ordinary capabilities. Blending Japanese-style continuous improvement with Silicon Valley culture, he approached the factory as if it were a software program (Teece, 2018). Tenacity, unwavering self-confidence, optimism, and a tolerance for stress are traits of successful entrepreneurs (Crane and Crane, 2007; Rauch and Frese, 2007). Passion for work, i.e., a willingness to put in the long hours for an extended period that a new venture requires, is another important feature (Baum and Locke, 2004). There are of course risks in relying on a particular talented individual, especially if those talents don’t translate into a set of replicable internal routines. It’s therefore important that an entrepreneurial state of mind and related skills are disseminated and encouraged throughout the organization. In 2008, before Steve Jobs’ second medical leave, he established an internal business school at Apple in which academics were brought in to prepare cases about how key past decisions, such as the creation of the Apple Store, were reached (Lashinsky, 2011). By having executives teach these cases to the company’s managers, Apple’s high-level routines and top management processes are propagated among its current and future leaders. Entrepreneurial thinking can also be embedded in the organization’s culture and repeatedly demonstrated and rewarded. This applies to dynamic capabilities more generally. The environmental scanning that supports “sensing,” for example, should be distributed throughout the organization, with open communication channels to the appropriate decision makers.

15.5

Entrepreneurial managers in the dynamic capabilities framework

Obviously, not everyone who starts a new enterprise wields the same abilities as Elon Musk or Steve Jobs. But there are many talented entrepreneurs, some quiet at their work, others quite noisy about it. Entrepreneurial managers also operate within established firms (Teece, 2007, 2016; Helfat and Peteraf, 2015). By “entrepreneurial manager,” I mean a particular combination of skills. Top and middle managers in an organization can be called on to deploy three types of roles described further in Table 15.1: operational, entrepreneurial, and leadership. These are each aspects of managerial dynamic capabilities (Adner and Helfat, 2003). The roles can be distributed in Table 15.1 Three Roles of Managers Operational Role RESPONSIBILITIES Planning and Budgeting ACTIVITIES

Organizing and Staffing

LEVERS

Control and Problem Solving Technical Efficiency and Predictable Results

GOALS

Entrepreneurial Role Sensing and Seizing

Leadership Role

Propagating Vision and Values Orchestrating Resources Aligning People with Strategy Investing in R&D, Developing Motivating New Business Models People Competitive Advantage Unity of Purpose

Source: Teece (2016).

204

Evolutionary economics, routines, and dynamic capabilities

many ways. For example, they may be explicitly divided across job titles or, in a small startup, they may be combined in a single individual. It is the combination of entrepreneurship and leadership that I call “entrepreneurial management.”3 This is not to imply that entrepreneurship within an established organization is the same as entrepreneurship in a new venture. The launch of a firm-within-a-firm, for example, places special challenges on top management’s ability to handle existing and new businesses “ambidextrously” (O’Reilly and Tushman, 2004, 2008). This requirement adds another layer of difficulty on the launch of a new business, and access to the resources of the established business can drag down the new one through “diseconomies of scope” (Bresnahan, Greenstein, and Henderson, 2011). As Bresnahan et al. point out in their detailed analysis of IBM’s missteps in the personal computer market and Microsoft’s in Internet infrastructure software, “the two incumbent firms had no difficulty building the raw organizational capabilities necessary to compete in the new markets” yet both were tripped up by “very considerable organizational conflict” between the new business and the old (Bresnahan, Greenstein, and Henderson, 2011, p. 266). Their inability to manage so as to allow both old and new businesses to advance marked a failure of top management’s transformation capabilities. Bresnahan et al.’s analysis also makes clear that strong organizational capabilities are not sufficient on their own to build and maintain competitive advantage. Strong dynamic capabilities for transforming the organization in a way that allows new, potentially incompatible initiatives to thrive are indispensable. Evolutionary economics has not entirely ignored the manager, arguing that the “intentionality and deliberation” implicit in the development of capabilities “provides a bridge between the predominantly descriptive concerns of evolutionary theory and the prescriptive analysis of firm strategy” (Dosi, Nelson, and Winter, 2000, p. 12). But the dynamic capabilities view makes the role of managers explicit, and more prominent. Sensing, seizing, and transforming, the main categories of dynamic capabilities, while reliant on organizational routines, are coordinated and guided by entrepreneurial managers. In the dynamic capabilities view, “the entrepreneur/manager function … is in part Schumpeterian (the entrepreneur introduces novelty and seeks new combinations) and in part evolutionary (the entrepreneur endeavors to promote and shape learning)” (Augier and Teece, 2009, p. 418). One could add that the entrepreneurial manager also leads transformation when that becomes necessary. Entrepreneurship and leadership are critical for the performance of the business enterprise and are difficult to teach. Whereas the operational role lends itself to routines and standard operating procedures, the entrepreneurial role is not readily amenable to routinization. Neither is leadership, which is required in order to unite people around a shared purpose and instill or sustain a culture of flexibility, innovation, and change. As Baumol (1968, p. 65) notes, leadership is virtually absent from the mainstream theory of the firm. It is also absent from evolutionary economics. In the dynamic capabilities view of the firm, entrepreneurial managers are central to the firm’s evolution because they have the ability to decide if existing capabilities will remain in the firm and whether new ones should be added. The continuity and evolution of the particular set of operational capabilities that a firm has at a point in time are important to the running of the firm, but they are of secondary importance for its long-term survival. Entrepreneurial management requires a rare mix of creativity and rigor: a vision of the future, a commitment to hypothesis testing, a love of risk, and sensitive people skills. 205

David J. Teece

Naturally not all managers bring the same level of skill to bear. Poor managers can hinder a firm’s development rather than propel it forward (Rosenbloom, 2000). There is, intentionally, a normative quality to the dynamic capabilities account of strategic management and economic change. Yet it is also positive, i.e., descriptive, in that all organizations do some sort of sensing, seizing, and transforming, even if they don’t do it well or in an organized fashion. The chief implication of the differing treatment of the entrepreneurial manager between dynamic capabilities and evolutionary economics is the relative emphasis on innovation and change. Major change can happen in evolutionary models, but it has a large random component. In the dynamic capabilities framework, it’s an objective. Nelson and Winter see dynamic capabilities and innovation as just so many routines: “we view firms as possessing routines which operate to modify over time various aspects of their operating characteristics” (Nelson and Winter, 1982, p. 17). But revolution cannot be readily routinized.

15.6

Dynamic capabilities: Evolution with design, purpose, and strategy

15.6.1 Contributions to dynamic capabilities by evolutionary economics Evolutionary thinking has been influential in strategic management and helped undergird the dynamic capabilities framework, particularly in its first iteration (Teece, Pisano, and Shuen, 1997). In that early version, I focused on the role of history in shaping the extent to which a firm can reconfigure its assets and concluded that a firm’s “evolutionary path … is often rather narrow” (p. 524). The initial definition of dynamic capabilities as “the firm’s ability to integrate, build, and reconfigure internal and external competences to address rapidly changing environments” (Teece, Pisano, and Shuen, 1997, p. 516) echoes the Nelson and Winter concept of “routines which operate to modify over time various aspects of [firms’] operating characteristics” (Nelson and Winter, 1982, p. 17). But, even in 1997, dynamic capabilities were already straining at the leash, being described further as “an organization’s ability to achieve new and innovative forms of competitive advantage given path dependencies and market positions” (Teece, Pisano, and Shuen, 1997, p. 516). Path dependencies are a context but not strictly a constraint. Dynamic capabilities has roots in evolutionary economics, but also goes beyond it. Both frameworks share common ancestors, drawing on the same behavioral economics underpinnings. The heritage of dynamic capabilities from the behavioral theory of the firm (Cyert and March, 1963; Gavetti, Greve, Levinthal, and Ocasio, 2012) includes “organizational expectations” and search (or “sensing,” in dynamic capabilities terms). The Cyert and March concepts of “organizational choice” and “organizational control” align with “seizing” in dynamic capabilities, and the behavioral notion of adaptation captures some elements of the dynamic capabilities for “transformation.” In general, where behaviorists (and evolutionists) emphasize limitations such as bounded rationality on human decisions, the dynamic capabilities framework emphasizes the possibility of devising new businesses as well as new futures for old businesses. To some extent, managerial epiphanies parallel scientific breakthroughs. As Einstein once noted: The mind can proceed only so far upon what it knows and can prove. There comes a point where the mind takes a leap—call it intuition or what you will—and comes out upon a

206

Evolutionary economics, routines, and dynamic capabilities

higher plane of knowledge, but can never prove how it got there. All great discoveries have involved such a leap. (cited in LIFE, 1955, p. 64) That’s not to say there isn’t a process involved, although it may not be a routine. Einstein also said that “intuition is nothing but the outcome of earlier intellectual experience” (cited in Isaacson, 2007, p. 113). Dynamic capabilities and evolutionary economics also each draw on concepts from Austrian economic thinking, particularly the work of Schumpeter, but also Kirzner, Hayek, and others. Teece is more Austrian than Nelson and Winter in the sense that the dynamic capabilities view embraces discontinuous change, emergent order, and “knowledge of the particular circumstances of time and place” (Hayek, 1945) as the norm in VUCA environments. Change does of course exist in evolutionary models of the firm, but it tends to be incremental and often unintentional. Firm capabilities can undergo drift, including depreciation, possibly due to employee turnover (Argote, 1996). Capabilities may be transferred and replicated, sometimes imperfectly (Teece, 1977; Winter, Szulanski, and Jensen, 2012). Such drift/mutation is an integral part of evolutionary theorizing. By contrast, large, deliberate step-function changes resulting from new combinations in the minds of entrepreneurial managers don’t fit all that easily into evolutionary thinking. Winter (2006) noted the difficulties this poses for building a model of the firm: Just as the fragmentation of knowledge in the firm makes innovation difficult and the consequences of attempted innovation unpredictable it tends to frustrate the economist who wants to predict the lines that innovation will take … . “Mere managers” may behave predictably, entrepreneurs (and the organizations led by them) do not. (Winter, 2006, p. 139) Subsequent to the earliest dynamic capabilities articles, I have taken a broader approach, constructing a dynamic capabilities framework that embraces the potential for firms to effectuate more fundamental, discontinuous organizational transformations (Teece, 2007, 2014). Entrepreneurial managers can search not just locally but widely for new opportunities and introduce routines more distant from existing ones than are typically contemplated in the evolutionary literature. Call it evolution with design, or even better, evolution with design, purpose, and strategy.

15.6.2

Narrow view of innovation and change in evolutionary economics

In evolutionary models, change is typically invoked with the word “innovation.” Purposeful discontinuities make organizations harder to model, but such innovations are the stuff of economic growth. Evolutionary innovations are generally modeled along a single dimension (usually technology) in which change occurs, when in fact there are many ways in which an organization can change, including its internal structure, organizational boundaries, business models, market positioning, and so on. Nelson has acknowledged as much:

207

David J. Teece

While most of the research by evolutionary economists, and scholars more broadly, on innovation has been oriented to technological innovations, increasingly organizational and institutional innovation is on the agenda. (Nelson, 2018, p. 24) The unidimensionality of innovation in evolutionary economics can be seen in NK models, a common tool in the discipline. The NK concept of distance is reductive because it compares sets of binary choices (e.g., Levinthal, 1997). In mathematical terms, it’s Hamming distance rather than Euclidean distance. It gives no insight into how difficult any individual change will be. Stuart and Podolny (1996) operationalized a quantitative metric for technological distance over an economic space of production techniques, but it’s not clear this can be generalized to other variables. In a recent theoretical paper (Teece, 2019a), I argued for the need to consider additional dimensions. I defined organizational change as occurring in a space dimensionalized by technology, markets, and business models (see Figure 15.1). The difficulty and, most likely, the cost of executing a shift to a new point in this space increases with distance from the firm’s existing capability set. While the assessments of the distances involved are necessarily qualitative, the structure forces decision makers to keep in mind the multidimensionality of the task at hand. I might well have added a dimension of time, because the pecuniary and other costs of making a significant change are likely to be inversely related to the amount of time over which the change must be made. The gravitational pull of the existing way of doing things makes it hard for new ways to achieve escape velocity—and most likely raises their implementation cost. The trade-off between the cost and speed of change can be mitigated to some extent by advanced preparation in the form of creating a culture of innovation and resilience. An innovative, agile culture cannot be created overnight. Like absorptive capacity, it builds over time and lowers the cost—and expands the range—of future strategic choices (Zahra and George, 2002). Conversely, the imposition of radical change in an organization that is not suitably prepared is likely to create problems that can potentially undermine strategic renewal (Teece, 2019b).

Market Distance Target state relative to current “O”

Current state

O

Technological Distance

Business Model Distance

Figure 15.1

The Dimensions of Distance for Transformation.

Source: Teece (2019a).

208

Evolutionary economics, routines, and dynamic capabilities

15.6.3 Dynamic capabilities and evolutionary economics reimagined As important as how firms can change is why they change. This is where dynamic capabilities can provide guidance. The key clusters of activities that make up dynamic capabilities can be categorized as sensing, seizing, and transforming (Teece, 2007, 2012). As explained above, dynamic capabilities are as dependent on managerial decisions as on organizational routines. As elastic as the concept of routines may be (Nelson and Winter, 1982, p. 115), the concept is not fully up to the tasks that the dynamic capabilities framework requires. The gap is most evident in sensing. The mainstream evolutionary equivalent of sensing is “search.” The tradition that has developed in evolutionary economics, particularly with respect to technology, has been that search is primarily “local.” Nelson and Winter (1982, p. 409) called for more research on “alternative” (i.e., non-local) strategies, but much of the work in the evolutionary field has privileged the pull of legacy over the push of strategy (Laursen, 2012). In the population-level NK models of Levinthal and colleagues, search is typically modeled in simulations as random changes in one or more dimensions (e.g., Levinthal and Posen, 2007). This allows comparisons of local and non-local search strategies, but at the expense of abstracting from how strategies are formed. By contrast, sensing, in the dynamic capabilities context, is the ability, under Knightian uncertainty, to either recognize opportunities before they are fully apparent or, in some cases, create new ones (Helfat and Peteraf, 2015). While there are underlying routines, the signals that feed into them should come from near and far, leaving it to the relevant decision maker(s) to make sense from them, as a prelude to making strategy. There is no obvious place for seizing in evolutionary economics. In the dynamic capabilities framework, seizing involves the innovation of business models, the filling of capability gaps, the achievement of alignment, the setting of firm boundaries, and the making of investment commitments. Aspects of these activities can be found by reading between the lines of the evolutionary literature, but they are certainly not given the full attention they merit in terms of their strategic importance. More importantly, evolutionary economics gives too little attention to the dimension of time, particularly the urgency needed for effective seizing. As for transforming, evolving organizations can adopt new routines; but the emphasis for evolutionists is on trial-and-error learning (Winter, 2000). The consequential decisions about what capabilities will be needed to execute a new business model and whether to make or buy those that are missing are exogenous. As the following quote from Jeff Bezos, Amazon CEO, indicates, entrepreneurial thinking and evolutionary thinking can be at cross purposes: Companies get skills-focused, instead of customer-needs focused … . A much more stable strategy is to start with “what do my customers need?” Then do an inventory of the gaps in your skills. Kindle is a great example. If we set our strategy by what our skills happen to be rather than by what our customers need, we never would have done it. We had to go out and hire people who know how to build hardware devices and create a whole new competency for the company. (Bezos, 2008) The Teecian view of dynamic capabilities allows proactive managers to effectuate organizational change in anticipation of environmental change instead of waiting to adapt

209

David J. Teece

after external changes occur. The development of firms is not by any means completely path dependent; nor is it limited to best-practice routines. Instead, distinctive, higher-order routines, rules of thumb, and/or managerial approaches lead to distinctive evolutionary paths. Excellence not only in search (“sensing” in dynamic capabilities terms) but also in sensemaking (Teece, 1998) affords the firm the opportunity to stay ahead of competitors. An overarching function of top management in the dynamic capabilities framework is “asset orchestration” (Teece, 2007, 2014).4 Orchestration encompasses not only routines and other resources of the firm, but also assets outside the firm, via strategic alliances, ecosystems, etc. Orchestration aims at alignment of all assets and, when successful, can itself be a source of value. Evolutionary theory is silent with respect to the asset structure of the firm and related alignment opportunities. Alignment is not routine; it requires artful coordination and continual attention. Orchestration and alignment also imply integration. As Winter has written, “the difficulties standing in the way of adaptive or innovative change … . have to do with the fragmentation of relevant knowledge, both of operations and concepts” (Winter, 2006, p. 138). The ability to integrate knowledge across large, or even smaller, organizations may come down more to their culture of internal openness and shared sense of purpose than to any identifiable routine. Culture and vision must, in turn, be fostered and reinforced by top management. Another significant component of the dynamic capabilities framework that is exogenous to evolutionary economics is strategy. Strategy-making processes are conceptually separate from dynamic capabilities; but both are core, interacting elements of the broader dynamic capabilities framework (Teece, 2014). Leveraging dynamic capabilities for competitive advantage, particularly in large organizations, requires good strategy, bold (and smart) decisions, and also leadership skills. These are all hard to routinize. The strategy element in the dynamic capabilities framework draws on Richard Rumelt’s strategy kernel, which holds that a good strategy generally involves a diagnosis (problem definition and analysis), a guiding policy (general approach adopted), and a coherent plan for actions to be taken (Rumelt, 2011). This sequence is slightly suggestive of a high-order routine, but to view it that way is to distract from the individual decisions that must ultimately be made. To the extent that strategy has been examined through an evolutionary lens, it has been largely a bottom-up model. Strategy is seen to originate with front-line managers and is then filtered and shaped by middle and top managers within the boundaries of the context set by top management (Noda and Bower, 1996). An internal selection process ensues within the firm as strategic initiatives compete for resources. The dynamic capabilities framework tends to view strategy with a top-down perspective, while still recognizing the critical roles that lower-level managers play in (1) collecting and communicating signals as inputs for managerial sensing and (2) translating and implementing strategy once it has been adopted (Lee and Teece, 2013). As with other aspects of the dynamic capabilities framework, the emphasis on what Winter (2017) calls System 2 decisions by top management are emphasized over the underlying System 1 process.

15.7

Conclusion

In this brief essay, I’ve discussed how evolutionary economics is one of the foundational pillars of dynamic capabilities. Another of these pillars is the entrepreneurial perspective, 210

Evolutionary economics, routines, and dynamic capabilities

which is more focused on the creation of discontinuities, i.e., the pioneering of new markets (Penrose, 1959; Denrell, Fang, and Winter, 2003). Each brings something different. While evolutionary models suggest adaptation of firms to their environment, the entrepreneurial approach promotes the ability of firms to shape that environment. The unification of these perspectives in the dynamic capabilities framework leads to a recognition of the ability/need to change while accounting for underlying continuities. It is my hope that, by viewing evolutionary economics within the broader context of the dynamic capabilities framework, evolutionary economists will be better able to handle a fuller range of innovations and non-routine entrepreneurial decisions. It may be time for evolutionary theorists to abandon the belief that routines or meta-routines can undergird the entirety of dynamic capabilities. Creative managerial and entrepreneurial acts (e.g., creating new markets), which drive economic growth and change, are, by their nature, strategic and nonroutine, even though there may be underlying principles that guide the choices.

Acknowledgments I would like to thank Greg Linden for research support and for many contributions and corrections. I am also grateful to Dick Nelson and Nicolas Petit for helpful comments.

Notes 1 Levinthal (2007) has observed that processes of path selection and resource allocation are underdeveloped in evolutionary economics. Resource allocation (part of asset orchestration) is a dynamic capability (Lovallo, Brown, Teece, and Bardolet, 2020). 2 I and my Industrial and Corporate Change co-editor Giovanni Dosi recognized the importance of this classic but unpublished RAND working paper and, with Winter’s concurrence, used our editorial authority, for the first time in the journal’s history, to have it published. 3 Arguably, I could have used “entrepreneurial management and leadership” to keep equal stress on insight and implementation. I chose to use “entrepreneurial management” for brevity. 4 This is perhaps better thought of as “wild orchestration” as compared to tightly choreographed coordination. In an orchestra, the conductor sticks to the score. The orchestration in dynamic capabilities is more akin to jazz, where there is a leader but all elements are also creating and responding in real time. As Reed Hastings, CEO of Netflix, put it in the conclusion to No Rules Rules (Hastings and Meyer, 2020): “To build a team that is innovative, fast, and flexible, keep things a little bit loose. Welcome constant change. Operate a little closer toward the edge of chaos. Don’t provide a musical score and build a symphonic orchestra. Work on creating those jazz conditions and hire the type of employees who long to be part of an improvisational band. When it all comes together, the music is beautiful.”

References Adler, P. (1993). The learning bureaucracy: New United Motor Manufacturing, Inc. Research in Organizational Behavior, 15, 111–194. Adler, P. S., Goldoftas, B., & Levine, D. I. (1999). Flexibility versus efficiency? A case study of model changeovers in the Toyota production system. Organization Science, 10(1), 43–68. Adner, R., & Helfat, C. E. (2003). Corporate effects and dynamic managerial capabilities. Strategic Management Journal, 24(10), 1011–1025. Argote, L. (1996). Organizational learning curves: Persistence, transfer and turnover. International Journal of Technology Management, 11(7–8), 759–769.

211

David J. Teece Arndt, F., & Pierce, L. (2018). The behavioral and evolutionary roots of dynamic capabilities. Industrial and Corporate Change, 27(2), 413–424. Augier, M., & Teece, D. J. (2009). Dynamic capabilities and the role of managers in business strategy and economic performance. Organization Science, 20(2), 410–421. Baum, J. R., & Locke, E. A. (2004). The relationship of entrepreneurial traits, skill, and motivation to subsequent venture growth. Journal of Applied Psychology, 89(4), 587. Baumol, W. J. (1968). Entrepreneurship in economic theory. American Economic Review, 58(2), 64–71. Becker, M. C., Lazaric, N., Nelson, R. R., & Winter, S. G. (2005). Applying organizational routines in understanding organizational change. Industrial and Corporate Change, 14(5), 775–791. Bezos, J. (2008). Bezos on innovation. Bloomberg BusinessWeek, April 16, 2008. https://www. bloomberg.com/news/articles/2008-04-16/bezos-on-innovation (accessed December 27, 2020). Bloom, N., Genakos, C., Sadun, R., & Van Reenen, J. (2012). Management practices across firms and countries. Academy of Management Perspectives, 26(1), 12–33. Bresnahan, T. F., Greenstein, S., & Henderson, R. M. (2011). Schumpeterian competition and diseconomies of scope: Illustrations from the histories of Microsoft and IBM. In J. Lerner, & S. Stern (eds.), The Rate and Direction of Inventive Activity Revisited. Chicago, IL: University of Chicago Press, 203–271. Burrows, P. (2004). The seed of Apple’s innovation. Businessweek.com, October 12. Retrieved from http://www.businessweek.com/print/bwdaily/dnflash/oct2004/nf20041012_4018_db083.htm (no longer available) Crane, F. G., & Crane, E. C. (2007). Dispositional optimism and entrepreneurial success. The Psychologist-Manager Journal, 10(1), 13–25. Cyert, R. M., & March, J. G. (1963). A Behavioral Theory of the Firm. Englewood Cliffs, NJ: PrenticeHall. Denrell, J., Fang, C., & Winter, S. G. (2003). The economics of strategic opportunity. Strategic Management Journal, 24(10), 977–990. Dosi, G., Nelson, R. R., & Winter, S. G. (2000). Introduction: The nature and dynamics of organizational capabilities. In G. Dosi, R. R. Nelson, & S. G. Winter (eds.), The Nature and Dynamics of Organizational Capabilities. Oxford, UK: Oxford University Press. Fruin, W. M., & Nishiguchi, T. (1990). The Toyota Production System: Its Organizational Definition in Japan. Economic Review, 42(1), 42–55. Gavetti, G., Greve, H. R., Levinthal, D. A., & Ocasio, W. (2012). The behavioral theory of the firm: Assessment and prospects. Academy of Management Annals, 6(1), 1–40. Gratton, L., & Ghoshal, S. (2005). Beyond best practice. MIT Sloan Management Review, 46(3), 49–57. Hastings, R., & Meyer, E. (2020). No Rules Rules: Netflix and the Culture of Reinvention. New York: Penguin Press. Hayek F. A. (1945). The use of knowledge in society. American Economic Review, 35(4), 519–530. Helfat, C. E. (2018). The Behavior and Capabilities of Firms. In R. R. Nelson, G. Dosi, C. E. Helfat, A. Pyka, S. G. Winter, P. P. Saviotti, K. Lee, F. Malerba, & K. Dopfer (eds.), Modern Evolutionary Economics: An Overview. Cambridge, UK: Cambridge University Press, 85–103. Helfat, C. E., & Peteraf, M. A. (2015). Managerial cognitive capabilities and the microfoundations of dynamic capabilities. Strategic Management Journal, 36(6), 831–850. Helfat, C. E., & Winter, S. G. (2011). Untangling dynamic and operational capabilities: Strategy for the (n)ever‐changing world. Strategic Management Journal, 32(11), 1243–1250. Hoopes, D. G., & Madsen, T. L. (2008). A capability-based view of competitive heterogeneity. Industrial and Corporate Change, 17(3), 393–426. Isaacson, W. (2007). Einstein: His Life and Universe. New York: Simon and Schuster. Kahneman, D. (2011). Thinking, Fast and Slow. New York: Farrar, Straus and Giroux. Kenney, M., & Florida, R. (1991). How Japanese industry is rebuilding the rust belt. Technology Review, 94(2), 24–33. Lashinsky, A. (2011). How Apple works: Inside the world’s biggest startup. Fortune (May 9, 2011). Available at https://fortune.com/2011/05/09/inside-apple/ (subscription required). Lee, S., & Teece, D. J. (2013). The functions of middle and top management in the dynamic capabilities framework. Kindai Management Review, 1, 28–40.

212

Evolutionary economics, routines, and dynamic capabilities Laursen, K. (2012). Keep searching and you’ll find: what do we know about variety creation through firms’ search activities for innovation? Industrial and Corporate Change, 21, 1181–122010.1093/icc/ dts025. Levinthal, D. A. (1997). Adaptation on rugged landscapes. Management Science, 43(7), 934–950. Levinthal, D. A. (2007). Bringing selection back into our evolutionary theories of innovation. In S. Brusoni, & F., Malerba (eds.), Perspectives on Innovation. Cambridge, UK: Cambridge University Press, 293–307. Levinthal, D., & Posen, H. E. (2007). Myopia of selection: Does organizational adaptation limit the efficacy of population selection? Administrative Science Quarterly, 52(4), 586–620. LIFE Magazine (1955). Death of a genius. May 2, 1955. 38(18), 61–64. Lippman, S. A., & Rumelt, R. P. (1982). Uncertain imitability: An analysis of interfirm differences in efficiency under competition. The Bell Journal of Economics, 418–438. Lovallo, D., Brown, A. L., Teece, D. J., & Bardolet, D. (2020). Resource re‐allocation capabilities in internal capital markets: The value of overcoming inertia. Strategic Management Journal, 41(8), 1365–1380. Nelson, R. R. (1991). Why do firms differ, and how does it matter? Strategic Management Journal, 12(S2), 61–74. Nelson, R. R., & Sampat, B. N. (2001). Making sense of institutions as a factor shaping economic performance. Revista de Economía Institucional, 3(5), 17–51. Nelson, R. R. (2018). Economics from an evolutionary perspective. In R. R. Nelson, G. Dosi, C. E. Helfat, A. Pyka, S. G. Winter, P. P. Saviotti, K. Lee, F. Malerba, & K. Dopfer (eds.), Modern Evolutionary Economics: An Overview. Cambridge, UK: Cambridge University Press, 1–34. Nelson, R. R., & Winter, S. G. (1982). An Evolutionary Theory of Economic Change. Cambridge, MA: Belknap Press. Noda, Tomo, & Bower, Joseph L. (1996). Strategy making as iterated processes of resource allocation. Strategic Management Journal, 17, 159–19210.1002/smj.4250171011. O’Reilly, C. A., & Tushman, M. L. (2004). The ambidextrous organization. Harvard Business Review, 82(4), 74–81. O’Reilly, C. A., & Tushman, M. L. (2008). Ambidexterity as a dynamic capability: resolving the innovator’s dilemma. Research in Organizational Behavior, 28, 185–206. Penrose, E. T. (1959). Theory of the Growth of the Firm. Oxford, UK: Blackwell. Popomaronis, T. (2020). Elon Musk calls this a ‘powerful, powerful way of thinking’—but is ‘hard to do.’ Here’s how it works. Blog dated February 29, 2020. https://www.cnbc.com/2020/02/28/ billionaire-elon-musk-this-is-a-powerful-way-of-thinking-but-hard-to-do-how-it-works.html (accessed January 5, 2021). Rauch, A., & Frese, M. (2007). Let’s put the person back into entrepreneurship research: A metaanalysis on the relationship between business owners’ personality traits, business creation, and success. European Journal of Work and Organizational Psychology, 16(4), 353–385. Rosenbloom, R. S. (2000). Leadership, capabilities, and technological change: The transformation of NCR in the electronic era. Strategic Management Journal, 21(10–11), 1083–1103. Rumelt, R. (2011). Good Strategy/Bad Strategy: The Difference and Why It Matters. New York: Crown Business. Schumpeter, J. A. (1934/1949). The Theory of Economic Development. Translated by R. Opie, Third Printing. Cambridge, MA: Harvard University Press. Stuart, T. E., & Podolny, J. M. (1996). Local search and the evolution of technological capabilities. Strategic Management Journal, 17(S1), 21–38. Teece, D. J. (1977). Technology transfer by multinational firms: The resource cost of transferring technological know-how. The Economic Journal, 87(346), 242–261. Teece, D. J. (1998). Capturing value from knowledge assets: The new economy, markets for knowhow, and intangible assets. California Management Review, 40(3), 55–79. Teece, D. J. (2007). Explicating dynamic capabilities: The nature and microfoundations of (sustainable) enterprise performance. Strategic Management Journal, 28(13), 1319–1350. Teece, D. J. (2012). Dynamic capabilities: Routines vs entrepreneurial action. Journal of Management Studies, 49(8), 1395–1401. Teece, D. J. (2014). The foundations of enterprise performance: Dynamic and ordinary capabilities in an (economic) theory of firms. Academy of Management Perspectives, 28(4), 328–352.

213

David J. Teece Teece, D. J. (2016). Dynamic capabilities and entrepreneurial management in large organizations: Toward a theory of the (entrepreneurial) firm. European Economic Review, 86, 202–216. Teece, D. J. (2018). Tesla and the reshaping of the auto industry. Management and Organization Review, 14(3), 501–512. Teece, D. J. (2019a). A capability theory of the firm: An economics and (strategic) management perspective. New Zealand Economic Papers, 53(1), 1–43. Teece, D. J. (2019b) Strategic renewal and dynamic capabilities: Managing uncertainty, irreversibilities, and congruence. In A. Tuncdogan, A. Lindgreen, F. Van Den Bosch, & H. Volberda (eds.), Strategic Renewal: Core Concepts, Antecedents, and Micro Foundations. Routledge: New York, 21–51. Teece, D., Peteraf, M., & Leih, S. (2016). Dynamic capabilities and organizational agility: Risk, uncertainty, and strategy in the innovation economy. California Management Review, 58(4), 13–35. Teece, D. J., Pisano, G., & Shuen, A. (1990). Firm capabilities, resources, and the concept of strategy. CCC Working Paper 90–8, Center for Research in Management, Berkeley: University of California. Teece, D. J., Pisano, G., & Shuen, A. (1997). Dynamic capabilities and strategic management. Strategic Management Journal, 18(7), 509–533. Urban, T. (2015). The cook and the chef: Musk’s secret sauce. Blog dated November 6, 2015. https:// waitbutwhy.com/2015/11/the-cook-and-the-chef-musks-secret-sauce.html#13 (accessed January 5, 2021). Winter, S. G. (2000). The satisficing principle in capability learning. Strategic Management Journal, 21(10–11), 981–996. Winter, S. G. (2003). Understanding dynamic capabilities. Strategic Management Journal, 24(10), 991–995. Winter, S. G. (2006). Toward a neo-Schumpeterian theory of the firm. Industrial and Corporate Change, 15(1), 125–141. Winter, S. G. (2017). Pursuing the evolutionary agenda in economics and management research. Cambridge Journal of economics, 41(3), 721–747. Winter, S. G., Szulanski, G., Ringov, D., & Jensen, R. J. (2012). Reproducing knowledge: Inaccurate replication and failure in franchise organizations. Organization Science, 23(3), 672–685. Womack, J. P., Jones, D. T., & Roos, D. (1990). The Machine That Changed the World. New York: Rawson Associates. Zahra, S. A., & George, G. (2002). Absorptive capacity: A review, reconceptualization, and extension. Academy of Management Review, 27(2), 185–203.

214

16 ROUTINES Markus C. Becker

16.1 The role of routines in evolutionary economics The three processes of variation, retention (inheritance) and selection are central in evolutionary theory.1 Understanding the mechanisms driving those three sub-processes is, therefore, crucial for developing evolutionary economics (Winter, 2014: 626). Candidates for selection and variation were easy to find, such as profit-induced selection processes among firms and the underlying selection mechanisms and innovation processes that generate variation in firms’ product portfolios. Before Winter’s work on the topic (Winter, 1964, 1971, 1975; Nelson & Winter, 1973, 1982), the missing element for a theory of economic evolution was a mechanism of retention (inheritance) (Becker & Knudsen, 2012). Without a mechanism of retention (inheritance), however, evolutionary economics could not explain how elements selected at one point of time were retained over time and available for being selected again at subsequent points of time. As Winter (1971) explained, in order for selection to operate effectively, there has to be a certain degree of stability. Winter (1975: 96) argued that a theory of natural selection must characterize the basic sources of continuity in the evolutionary process. In the biological case, this basic source is the genetic transmission of characteristics. If there were no causal link between the characteristics of the n-th generation and the characteristics of the n+1st, there could be no natural selection and no evolution. There would be no ‘descent with modification’, in Darwin’s famous phrase. An explanation of the ‘descent’ part would be missing. Nelson and Winter suggested possible candidates for sources of continuity and ‘units of selection and inheritance, that is, the ‘routine application of established rules, procedures, and policies’ (1982: 240–41), where ‘the decision rules themselves are the economic counterpart of genetic inheritance (1982: 245)’ (Becker & Knudsen, 2012: 244). They also write: Our general term for all regular and predictable behavioral patterns of firms is ‘routine’ … In our evolutionary theory, these routines play the role that genes play in biological

DOI: 10.4324/9780429398971-18

215

Markus C. Becker

evolutionary theory. They are a persistent feature of the organism and determine its possible behavior (though actual behavior is determined also by the environment); they are heritable in the sense that tomorrow’s organisms generated from today’s have much of the same characteristics, and they are selectable in the sense that organisms with certain routines may do better than others, and, if so, their relative importance in the population (industry) is augmented over time. (Nelson & Winter, 1982: 14) Nelson and Winter’s key idea was that ‘routines embody behavioral continuity of firms’ (Becker & Knudsen, 2012: 245), and ‘respond to the theoretical need for an inheritance mechanism in an evolutionary theory’ (Winter, 2014: 630). By supplying the missing link, this idea became ‘one of the pillars of evolutionary economics’ (Becker & Knudsen, 2012: 245). The ‘proposition that the behavior of firms at any time is to a considerable extent determined by the relatively automatic implementation of a set of routines’ also ‘remains a hallmark feature of the theory of the firm employed by evolutionary economists’ (Nelson, 2020: 5). Routines are the missing ‘key source of the continuity in behavior’ that is required if ‘ways of doing things’ are to be shaped by a truly evolutionary process’ (Winter, 2005a: 39–40; Winter, 1990; Knudsen, 2008). Because organizational routines are fundamental sources of persistence in organizational features, they possess the stability that selection requires (Knudsen, 2008: 143). Thus, routines have a fundamental role in evolutionary theories of cultural and economic change (Knudsen, 2008) and enable an evolutionary explanation of how work is done in organizations and how that changes over time. With this idea, Nelson and Winter (1982) introduced ‘a perspective on organizations that views their behavioral repertoires as bundles of routines’ (Becker & Knudsen, 2012: 248). From an evolutionary perspective, which is interested in the population level, recognizing organizational routines as units of analysis thus suggests a focus on populations of organizational routines. It also raises the question what mechanisms drive the variation, selection, and retention (inheritance) of routines.

16.2 16.2.1

Selection

Beyond genotypes and phenotypes: Routines as replicators and firms as interactors

The economy and the social world more generally are, obviously, different from the biological realm. Parallels and differences between evolution in different realms have been a matter of much debate. To avoid confusion, it seems helpful to leverage an insight from prior research. While ‘routines as genes’ was shorthand for pointing in a certain direction, an evolutionary theory of economic and organizational change does not have to rely on – and be constrained by – metaphor or analogy to biology. Rather, research has identified principles – variation, selection and continuity – common to all evolving systems (Hodgson & Knudsen, 2010). They are, in this sense, general features of evolving systems rather than features of the biological realm. It is important to note, however, that the content of these principles is specific to each domain of evolution (Knudsen, 2004: 150; Hodgson & Knudsen, 2010). These principles are, therefore, not specific to biological organisms but are abstract features of an evolutionary explanation (e.g., Price’s (1995) generalized concept of selection, Knudsen, 2004, 2008; Hodgson & Knudsen, 2010). An important step in this endeavor was Hull’s (1981) characterization of the abstract roles that genes and phenotypes

216

Routines

have in evolutionary theory (Hodgson & Knudsen, 2004). Hull pointed out that genes are specific instances of replicators, while phenotypes are specific instances of interactors. Replicators are ‘entities that pass on their structure directly in replication’ (Hull, 1981: 150). Interactors are ‘entities that produce differential replication by means of directly interacting as cohesive wholes with their environments’ (Hull, 1981: 150). Replicators play a central role in retention processes, while interactors play a central role in selection processes. As explained by Knudsen (2004: 161), powerful arguments support ‘the necessity of the replicator/interactor distinction in any form of enduring evolutionary process, and empirical evidence supports the actual existence of this distinction in economic and organizational evolution’. In biological organisms, genes serve the role of replicators. The concepts of replicator and interactor are two very useful concepts for understanding the selection and retention of routines. To fulfill a role in retention, routines need to pass on their structure directly in replication. They do not need to be like genes, which are one specific instantiation of replicators in biological organisms. Routines fulfil the role of replicators in evolutionary economic theory (Knudsen, 2004, 2008). The same applies to interactors (‘entities that produce differential replication by means of directly interacting as cohesive wholes with their environments’, Hull, 1981: 150). In biological organisms, genes do not interact directly with their environment. The organisms that carry them do. These organisms are selected, and with them the genes they carry, leading to differential replication of genes. Routines do not directly interact as cohesive wholes with their environments either; firms do (Hodgson & Knudsen, 2004b, 2010). Firms comprise routines, and their actions are shaped by their routines. The performance of firms, which is subject to selection pressure, is shaped by routines, too. Firms are thus considered interactors (Hodgson & Knudsen, 2004b, 2010). As Hodgson and Knudsen (2010) explain: The firm is not simply an aggregate of individuals, physical capital and codifiable knowledge. It also consists of idiosyncratic structures, relationships and routines that typically are not readily tradable and are specific to the firm itself (Winter, 1988; Langlois & Robertson, 1995). These routines are important repositories of knowledge that is not readily codified and sold. This means that most or all of the firms’ routines share the fate of the firm in which they reside. The competitive selection of cohesive groups such as firms is due to their differential properties in a common environment. In turn, these differential properties of firms partly emanate from the organized structure of the firms as a whole, and are not merely due to the aggregate properties of the individuals in the firm, taken severally. Structured and cohesive interactions between individuals within the firm give rise to, and are properly regarded as, properties of the firm. These are a cause of differential profitability and thus differential replication of the firm’s routines, i.e., competitive selection. (Hodgson & Knudsen, 2010: 173–4) Firms are interactors that comprise routines, which are replicators. Thus, one way in which routines get selected is through ‘contested market selection’ of firms (Levinthal, 2017: 285). As successful innovators expand and unprofitable firms are eliminated (Nelson, 2020: 6), routines are selected for through the selection of firms (Sober, 1984; Knudsen, 2002, 2004). 217

Markus C. Becker

16.2.2

Selection of firms and selection of routines: ‘Natural’ selection and managerial (‘artificial’) selection

Selection for routines by selection of firms on markets is not the only way in which routines are selected for. Executives in firms also consciously select among alternative possibilities (Levinthal, 2017: 285). As Nelson (2020: 9) notes, ‘much of the selection process proceeds through conscious decision making by economic actors regarding what to do’. Managers can choose directly between different ways of accomplishing a task, e.g., for doing quality management, manufacturing, or sales; they can choose different people or teams who have particular skills and competences in doing the task, including tacit knowledge (Knudsen, 2002); they can also choose key success factors, metrics and milestones (Levinthal, 2017, 2021). Thus, they define selection criteria and constitute ‘an artificial selection environment that guides the cultivation’ of specific routines in the organization (Levinthal, 2017: 283). This possibility of directly selecting for routines and changing the population of routines in an organization in this way represents an important contrast to biology. There, ‘phenotypes are stuck with the genes they are born with’ (Nelson, 2020: 6). In biology, there is no selection of individual genes, only selection of the organisms that carry them and in this way, selection for those genes (Sober, 1984; Hodgson & Knudsen, 2004b, 2010; Knudsen, 2002, 2004). Empirical studies (e.g., Bloom & Van Reenen, 2007, Syverson, 2011) on the impact of management show that such an internal selection environment is an important factor shaping the heterogeneity of routines in an industry (Winter, 2014: 623). The possibility of consciously shaping the internal selection environment allows integrating conscious decision-making by individual agents with evolutionary dynamics, in a way that acknowledges the influence of both in the realm of the economy (Levinthal, 2017).

16.3 16.3.1

Retention (inheritance) What is being retained?

To capture how routines are replicated, it is useful to have a clear view of what a routine is and thus, of what is being retained. As it turns out, different ideas can be found in the literature. They seem to point in different directions. As pointed out by Becker (2004), Becker and Knudsen (2012) and Becker (2016), scholars have considered organizational routines to be (1) repeated behavior patterns for accomplishing tasks (e.g., Pentland & Rueter, 1994; Pentland, 2003a, 2003b), (2) standard operating procedures or rules (Cyert & March, 1963; March et al., 2000), or (3) dispositions to engage in previously adopted or acquired behavior, triggered by an appropriate stimulus or context (Hodgson & Knudsen, 2004a, 2004b; Cohen, 2007). The last group of scholars ‘emphasize that routines as well as (stable) rules are stored behavioral capacities or capabilities, while their expression can be observed as actual patterns of behavior. These capacities involve knowledge and memory. They involve organizational structures and individual habits that, when triggered, lead to sequential behaviors (Hodgson & Knudsen, 2004a, 2004b)’ (Becker & Knudsen, 2012: 245; Becker, 2004, 2016). While these three ideas of routines appear contradictory, they fit neatly into an overall conception of organizational routines: The concept of organizational routines provides a perspective on organizations that focuses on regularities and stability in how organizational tasks are accomplished 218

Routines

jointly by several actors. Such a perspective provides a concept for describing stable multi-person behavior (recurrent interaction patterns), and two different potential causes of the stability of such behavior (standard operating procedures and dispositions). Standard operating procedures are an instance of conventions while dispositions are an instance of individual-level causes. The two potential causes of stable behavior therefore comprise both macro- (top-down) and micro- (bottom-up) causes. The three perspectives on behavioral continuity jointly offer a perspective on the sources and nature of the pool of stable repertoires that characterize firm behavior. (Becker & Knudsen, 2012: 245; Becker, 2016) A key insight for understanding the replication of routines is that organizational routines are generative: They refer to what generates behavior, not to the behavior itself (Hodgson & Knudsen, 2010). This is perhaps seen most clearly when considering that individual-level habits are important building blocks of organizational routines. Importantly, current research in psychology defines habits as ‘behavioral dispositions to repeat well-practiced actions given recurring circumstances’ (Wood, Tam & Witt, 2005: 918). That is, habits are generative, rather than behavior (Ouellette & Wood, 1998; Wood, Tam & Witt, 2005; Wood & Rünger, 2016; Wood & Neal, 2009). Nelson and Winter (1982, ch. 4) already suggested individual skills as a micro-foundation of the organizational-level analysis (ch. 5), which centered around organizational routines and thus, pointed to habit as a micro-foundation for routines (Winter, 2017). In the 1990s, research in psychology identified procedural memory (‘skill memory’) (Singley & Anderson, 1989) as the mechanism by which individuals remember how to do things. Leveraging this concept from psychology, Cohen and Bacdayan (1994) showed in an experimental study how organizational routines are stored in the procedural memory of individuals. Thereby, they ‘sketche[d] an explanatory bridge that runs all the way from brain physiology at one end – the physical locus of procedural memory – to the capabilities of large organizations at the other. The experiment specifically supports the central span in that bridge, which links the well-practiced behavior of individuals (skill) to the wellpracticed and coordinated behavior of a group (i.e., organizational capability, in this case that of a dyad)’ (Winter, 2017: 733–4). Building on these elements and on current research on habits in psychology (e.g., Ouellette & Wood, 1998; Wood, Tam & Witt, 2005; Wood & Rünger, 2016; Wood & Neal, 2009), Winter and others concluded that ‘a suitable individual-level foundation [of organizational routines] can be found only in an account of individual psychology that gives due weight to habit and clearly distinguishes habit from deliberative decision making’ (Winter, 2013: 120). Likewise, Cohen, Levinthal, and Warglien (2014: 331) argue that formal models of routines ‘must be consistent with the psychological processes of actors whose actions are determined in large part by learned habits and associations rather than by deliberating over the likely consequences of exogenously defined alternatives. … it is these habit-based processes that give routines … their distinctive common qualities’. Routines can, thus, be considered ‘organizational meta-habits, existing on a substrate of habituated individuals in a social structure’ (Hodgson & Knudsen, 2010: 171). They ‘depend upon a structured group of habituated individuals […]. The behavioral cues by some members of a structured assembly of habituated individuals trigger specific habits in others. Hence, various individual habits sustain each other in an interlocking structure of reciprocating individual behaviors’ (Hodgson & Knudsen, 2010: 141). 219

Markus C. Becker

It is important to note that both in classic (Dewey, 1922) and current psychology, habits are defined not as stable behavior patterns of individuals but rather, as ‘behavioral dispositions to repeat well-practiced responses as they are triggered automatically by recurring circumstances’ (Wood, Tam & Witt, 2005: 918; Ouellette & Wood, 1998). Habits are not stable behavior patterns of individuals, but what generates them (Hodgson & Knudsen, 2010: 239; Birnholtz, Cohen & Hoch, 2007; Cohen, 2007). Reflecting their generative nature, organizational routines are, therefore, ‘organizational dispositions to energize conditional patterns of behavior within organizations, involving sequential responses to cues that are partly dependent on social positions in the organization’ (Hodgson & Knudsen, 2010: 346).

16.3.2

The retention (inheritance) of routines

After arguing that the unit of selection needs to have stability, and pointing to organizational routines as possible units of selection, Winter’s third pioneering insight regarding routines was to identify a mechanism of retention (inheritance), i.e., the replication of organizational routines (Becker & Knudsen, 2012): In economic evolutionary theory, inheritance is first of all a matter of replication of successful routines by the individual organization. (Winter, 1990: 281) Nelson and Winter (1982: 118–124) already pointed out that routines as building blocks of organizational capabilities capture tacit knowledge. With regard to the ability to replicate a routine, this provides an advantage to being able to observe a working ‘template’ of a routine in action, while it establishes a barrier to imitation for other firms (Winter, 1995; Winter, 2005b). With colleagues, Winter (1990; 1995; Baden-Fuller & Winter, 2007) and Szulanski (Winter & Szulanski, 2001; Szulanski & Jensen, 2004; Jensen & Szulanski, 2007) investigated the replication of routines in firms (by the original firm) and across organizations (imitation), supporting the idea that having full access to the template matters for successful replication because it enables resolving problems in the copy by closer scrutiny of a working original (Winter & Szulanski, 2001; Winter, 2005b). The clarification that routines are generative leads to additional insights on the replication of routines. First, if routines are generative, replicating them means to replicate what generates the behavior patterns, rather than the behavior patterns. Behavior patterns might, for instance, also be stable because of exogenous constraints. If routines are behavioral dispositions, then replicating routines means to replicate behavioral dispositions. As Hodgson and Knudsen (2010: 49) explain, inheritance ‘must involve transmission from some kind of genotype’ – i.e., a behavioral disposition – ‘to another of the same kind’. Second, how does such replication of behavioral dispositions happen? While it is not the behavior patterns that are replicated, behavior patterns do play an important role in the replication of routines: behavioral dispositions replicate via their behavioral expressions: Unlike the replication of DNA or computer viruses, habits do not directly make copies of themselves. Instead they replicate indirectly, by means of their behavioral expressions. They can impel behavior that is consciously or unconsciously followed by others, as a result of incentive or imitation. It is possible, but not always necessary, 220

Routines

that codifiable rules or instructions are also involved. Eventually, the copied behavior becomes rooted in the habits of the follower, thus transmitting from individual to individual an imperfect copy of each habit by an indirect route. (Hodgson & Knudsen, 2004b: 287) Third, it matters for the replication of routines that organizational routines are more than habits, i.e., they are ‘organizational dispositions to energize conditional patterns of behavior within organizations, involving sequential responses to cues that are partly dependent on social positions in the organization’ (Hodgson & Knudsen, 2010: 140). As Winter (2005b: 238) explains, productive knowledge ‘is crucially a matter of distributed knowledge – i.e., of complementary parts of the same functional unit of knowledge being held by several individuals and applied in a coordinated, way, the coordination itself being a crucial aspect of the knowledge’. Social positions and the cues they provide play an important role in providing coordination. Accordingly, replicating routines involves both the replication of the organizational dispositions and of the ‘social positions that define legitimate roles associated with the relevant individual interactions and the performance of the routine’ (Hodgson & Knudsen, 2010: 142; also see Winter, 1995).

16.4

Variation of routines

The third main pillar of evolutionary theory consists in mechanisms driving the existence and replenishment of variety. For evolutionary economics, which considers organizations to be bundles of routines (Nelson & Winter, 1982), explaining variation of routines is a key question: As Veblen said, “For the purpose of economic science, the process of cumulative change that is to be accounted for is the sequence of change in the methods of doing thing(s) — the methods of dealing with the material means of life” (Veblen, 1898: 10). That was the program; that is the program. (Winter, 2014: 625-6 [emphasis in the Veblen quote added by the author]; also see Winter, 1986, 1990). Like for selection, there are obvious candidates for sources of variation: innovation (or invention) but also unintended consequences of conscious effort (Becker & Knudsen, 2012). At first blush, explaining how variety is generated in routine-based organizations (Nelson & Winter, 1982) is not obvious, however (Becker & Knudsen, 2012: 248; Becker, Knudsen & March, 2006). Prior research has identified several types of change that need to be accommodated in an evolutionary theory of the firm: ‘(1) incremental changes in vital properties of existing routines; (2) inter-firm and intra-firm diffusion of routines; and (3) the generation of distinctively novel routines.’ (Becker & Knudsen, 2012: 248–9). Learning theories can explain the first, and diffusion theories the second (Becker, Knudsen & March, 2006). To explain the third, the combinatorics of routines and unreliable routine replication have been identified as two possible sources of new variation in economic evolution (Becker, Knudsen & March, 2006). Combining extant elements, such as routines, in different ways opens a combinatorial space that offers great potential for variation, even if the elements that are combined are extant routines. Unreliable routine replication can be ‘a source of a random walk in the realized performance of organizations’ (Becker & Knudsen, 2012: 249) and thus, introduce variation. The unreliable replication of information in the replication 221

Markus C. Becker

process can also introduce variation, which might be picked up in the next round of replication (Zollo & Winter, 2002; Hodgson & Knudsen, 2010). A whole body of research on ‘routine dynamics’ (Feldman & Pentland, 2003; Feldman et al., 2016, 2021) has highlighted that because routines are effortful and emergent accomplishments, routines can change endogenously over iterations; in other words, the replication of routines over time is bound to be unreliable. Research on routine dynamics has shed light on many ways in which routines can change as they are replicated in subsequent time periods. Those include the interaction between the practiced objective and subjective dimensions of routines and between the context routines are embedded in and the agency of those who are involved in the routine, as well as with the artifacts involved in routines (Feldman et al., 2016; 2021). Furthermore, conflict can also be a source of variation in routines, either of incremental changes or of distinctively novel routines. Nelson and Winter’s (1982) notion of routines as representing a truce implies that as a truce concerning how to do a particular task unravels, pressure for changing the routine might ensue. Intra-organizational conflict, driven amongst others by processes of intra-organizational politics, appear to be important sources of the unravelling of truces (Zbaracki & Bergen, 2010; Kaplan, 2015; Winter, 2017; Salvato & Rerup, 2018). For this reason, it might be illuminating to understand ‘how the trends arising from the prevailing routines generate crises and choice situations, from which novelty – of a broadly predictable kind – could possibly emerge’ (Winter, 2014: 614).

16.5 Variation, selection, and retention, and their interaction The interaction of interactors with the environment can drive evolution in different ways. Theories of economic evolution have taken two different perspectives: A first perspective views environmental feedback ‘as a selection force causing differential elimination of organizations that hold inertial routines’, a second one, ‘as a force adjusting the variation among behavioral repertoires, which in turn forms the basis of internal adaptation and selection processes that change organizational routines’ (Knudsen, 2008: 143). That is, environmental feedback can select organizations that carry certain routines – for instance, as customers cease to buy a firm’s product – and thereby adjust the population of routines, or impact the adaptation of individual routines, e.g., when firm adapt their market research and advertising routines to better address relevant customer segments. Prior research has pointed to interdependencies between the processes of selection and adaptation (e.g., Levinthal, 1991; Levinthal & Posen 2007). As central pillars of an evolutionary theory of economic change, organizational routines are also implicated in those interdependencies between selection and adaptation: variation of routines impacts the outcomes of selection processes that are relevant for the population of organizational routines, and selection – as well as replication – of routines impacts the subsequent variation of routines in that population (Levinthal, 2021).

16.6

Conclusion

Organizational routines have central roles in evolutionary economic theory. Most importantly, they are replicators, i.e., ‘entities that pass on their structure directly in replication’ (Hull, 1981: 150) and thus, play a key role in retention. With this suggestion, Nelson and Winter (1982) filled one of the main gaps in evolutionary economic theory at the time. Their proposal was path-breaking and represented a huge step forward for evolutionary economic 222

Routines

theory. With increasing understanding and development of theory on economic evolution and organizational routines, over the last four decades, progress has been made concerning the variation, selection and retention of routines and their impact on variation, selection and retention of organizations. Yet, developments in evolutionary theory have also raised new questions about the role of routines in theory on economic evolution and the notion of routines as the equivalent of the gene (e.g., Levinthal, 2021). While organizational routines continue to be a central pillar for the development of evolutionary economic theory, many questions still remain open, calling for more research.

Note 1 This chapter draws on material from Becker and Knudsen (2012) and Becker (2016).

Bibliography Baden-Fuller, Charles, and Sidney G. Winter (2007): Replicating Organizational Knowledge: Principles or Templates? SSRN Working Paper, SSRN ID 1118013. Becker, Markus C. (2004): Organizational routines: A review of the literature. Industrial and Corporate Change. Vol. 13, No. 4, 643–677. Becker, Markus C. (2016): Organizational Routines. Teece, David, and Mie Augier (eds.): The Palgrave Dictionary of Strategic Management. London: MacMillan. Becker, Markus C., and Thorbjørn Knudsen (2012): Nelson and Winter Revisited. Dietrich, Michael, and Jackie Kraft (eds.): Handbook on the Economics and Theory of the Firm. Cheltenham: Edward Elgar, 243–255. Becker, Markus C., Thorbjørn Knudsen, and James G. March (2006): Schumpeter, Winter, and the sources of novelty. Industrial and Corporate Change, Vol. 15, No. 2, 353–371. Birnholtz, Jeremy P., Michael D. Cohen, and Susannah V. Hoch (2007): Organizational character: On the regeneration of Camp Poplar Grove. Organization Science, Vol. 18, No. 2: 315–332. Bloom, Nicholas, and John Van Reenen (2007): Measuring and explaining management practices across firms and countries. Quarterly Journal of Economics, Vol. CXXII, No. 4: 1351–1408. Cohen, Michael D. (2007): Reading Dewey: Reflections on the study of routine. Organization Studies, Vol. 28, No. 5, 773–786. Cohen, Michael, and Paul Bacdayan (1994): Organizational routines are stored as procedural memory: Evidence from a laboratory study. Organization Science, Vol. 5, No. 4, 1994, 554–568. Cohen, Michael D., Daniel A. Levinthal, and Massimo Warglien (2014): Collective performance: Modeling the interaction of habit-based actions. Industrial and Corporate Change, Vol. 23, No. 2, 329–360. Cyert, Richard M., and James G. March (1963): A Behavioral Theory of the Firm. Blackwell, Oxford. Dewey, John (1922): Human Nature and Conduct – An Introduction into Social Psychology. London: George Allen & Unwin. Feldman, Martha S., and Brian T. Pentland (2003): Reconceptualizing organizational routines as a source of flexibility and change. Administrative Science Quarterly, Vol. 48, No. 1: 94–118. Feldman, Martha S., Brian T. Pentland, Luciana D’Adderio, Nathalie Lazaric (2016): Beyond routines as things: Introduction to the special issue on routine dynamics. Organization Science, Vol. 27, No. 3: 505–513. Feldman, Martha S., Brian T. Pentland, Luciana D’Adderio, Katharina Dittrich, Claus Rerup, and David Seidl (2021): What Is Routine Dynamics? in Feldman, Martha S., Brian T. Pentland, Luciana D’Adderio, Katharina Dittrich, Claus Rerup, and David Seidl (eds.): Handbook of Routine Dynamics. Cambridge: Cambridge University Press, ch. 1. Hodgson, Geoffrey M., and Thorbjørn Knudsen (2004a): The complex evolution of a simple traffic convention: The functions and implications of habit. Journal of Economic Behavior & Organization, Vol. 54, No. 1: 19–47.

223

Markus C. Becker Hodgson, Geoffrey M., and Thorbjørn Knudsen (2004b): The firm as an interactor: Firms as vehicles for habits and routines. Journal of Evolutionary Economics, Vol. 14, No. 3: 281–307. Hodgson, Geoffrey M., and Thorbjørn Knudsen (2006): Dismantling Lamarckism: Why descriptions of socio-economic evolution as Lamarckian are misleading. Journal of Evolutionary Economics, Vol 16: 343–366. Hodgson, Geoffrey M., and Thorbjørn Knudsen (2010): Darwin’s Conjecture: The Search for General Principles of Social and Economic Evolution. Chicago: University of Chicago Press. Hull, David (1981): Units of Evolution: A Metaphysical Essay. (Asquith, P., and T. Nickles (eds.): Philosophy of Science Association, East Lansing/MI, 1:315-322. reprinted in: Brandon, Robert N./ Burian, Richard M. (eds.): Genes, Organisms, Populations – Controversies over the Units of Selection. MIT Press, Cambridge/MA, 142-160. Jensen, Robert J., and Gabriel Szulanski (2007): Template use and the effectiveness of knowledge transfer. Management Science, Vol. 53, No. 11: 1716–1730. Kaplan, Sarah (2015): Truce breaking and remaking: The CEO’s role in changing organizational routines. Advances in Strategic Management, Vol. 32, 1–45. Knudsen, Thorbjørn (2002): Economic selection theory. Journal of Evolutionary Economics, Vol. 12: 443–470. Knudsen, Thorbjørn (2004): General selection theory and economic evolution: The Price equation and the replicator/interactor distinction. Journal of Economic Methodology, Vol. 11, No. 2: 147–173. Knudsen, Thorbjørn (2008): Organizational Routines in Evolutionary Theory. In Becker, Markus C. (ed.): Handbook of Organizational Routines. Cheltenham: Edward Elgar: 125–151. Langlois, Richard N., and Paul L. Robertson (1995): Firms, Markets and Economic Change. London: Routledge. Levinthal, Daniel A. (1991): Organizational adaptation and environmental selection – Interrelated processes of change. Organization Science, Vol. 2, No. 1, 140–146. Levinthal, Daniel A. (2017): Mendel in the C-suite: Design and the evolution of strategies. Strategy Science, Vol. 2, No. 4: 282–287. Levinthal, Daniel A., and Hart E. Posen (2007): Myopia of selection: Does organizational adaptation limit the efficacy of population selection? Administrative Science Quarterly, Vol. 52, No. 4: 586–620. Levinthal, Daniel A. (2021). Evolutionary Processes and organizational adaptation. Oxford: Oxford University Press. March, James G., Martin Schulz and Xueguang Zhou (2000): The Dynamics of Rules – Change in Written Organizational Codes. Stanford: Stanford University Press. Nelson, Richard R. (2020): A perspective on the evolution of evolutionary economics. Industrial and Corporate Change, 1–18. Nelson, Richard, and Sidney G. Winter (1973): Toward an evolutionary theory of economic capabilities. American Economic Review (Papers and Proceedings), Vol. 68, No. 2: 440–449. Nelson, Richard, and Sidney Winter (1982): An Evolutionary Theory of Economic Change. Cambridge/ MA: Belknap Press of Harvard University Press. Ouellette, Judith A., and Wendy Wood (1998): Habit and intention in everyday life: The multiple processes by which past behavior predicts future behavior. Psychological Bulletin, Vol. 124, No. 1: 54–74. Pentland, Brian T. (2003a): Conceptualizing and measuring variety in the execution of organizational work processes. Management Science, Vol. 49, No. 7: 857–870. Pentland, Brian T. (2003b): Sequential variety in work processes. Organization Science, Vol. 14, No. 3: 528–540. Pentland, Brian and Rueter, Henry (1994): Organizational routines as grammars of action. Administrative Science Quarterly, Vol. 39: 484–510. Price, George R. (1995): The nature of selection’. Journal of Theoretical Biology, Vol. 175: 89–396. Salvato, Carlo, and Claus Rerup (2018): Routine regulation: Balancing conflicting goals in organizational routines. Administrative Science Quarterly, Vol. 63, No. 1: 170–209. Singley, Mark K., and John R. Anderson (1989): The Transfer of Cognitive Skill. Cambridge/MA: Harvard University Press. Sober, Elliott (1984): The Nature of Selection: Evolutionary Theory in Philosophical Focus. Cambridge, MA: MIT Press. Syverson, Chad (2011): What determines productivity? Journal of Economic Literature 2011, Vol. 49, No. 2: 326–365.

224

Routines Szulanski, Gabriel, and Robert J. Jensen (2004): Overcoming stickiness: An empirical investigation of the role of the template in the replication of organizational routines. Managerial & Decision Economics, Vol. 25, No. 6–7: 347–363. Veblen, Thorstein (1898): Why is economics not an evolutionary science? Quarterly Journal of Economics, Vol. 12: 373–397. Winter, Sidney G. (1964): Economic ‘natural selection’ and the theory of the firm’, Yale Economic Essays, Vol. 4 (Spring 1964): 225–272. Winter, Sidney G. (1971): Satisficing, selection and the innovating remnant. Quarterly Journal of Economics, Vol. 85 (May 1971): 237–261. . Winter, Sidney G. (1975): Optimization and Evolution in the Theory of the Firm. In Day, R. H., and T. Groves (eds.): Adaptive Economic Models. Academic Press, 73–118. Winter, Sidney G. (1986): The Research Program of the Behavioral Theory of the Firm: Orthodox Critique and Evolutionary Perspective. In Gilad, B., and S. Kalish (eds.): Handbook of Behavioral Economics, Vol. 1, Greenwich, CT: JAI Press, 151–188. Winter, Sidney G. (1988): On coase, competence, and the corporation. Journal of Law, Economics, and Organization, Vol. 4, No. 1: 163–180. Winter, Sidney G. (1990): Survival, Selection, and Inheritance in Evolutionary Theories of Organization. Singh, Jitendra V. (ed.): Organizational Evolution – New Directions. Newbury Park/ CA: Sage, 269–297. Winter, Sidney G. (1995): Four Rs of Profitability: Rents, resources, routines, and replication. Montgomery, Cynthia (ed.): Resource-based and evolutionary theories of the firm - Towards a synthesis. Dordrecht: Kluwer, 147–178. Winter, Sidney G. (2005a): Developing Evolutionary Theory for Economics and Management’ in Hitt, M., and K. G. Smith (eds.): Great Minds in Management: The Process of Theory Development, Oxford: Oxford University Press, 510–547. Winter, Sidney G. (2005b): Towards an Evolutionary Theory of Production. In Dopfer, Kurt (ed.): The Evolutionary Foundations of Economics. Cambridge: Cambridge University Press, 223–254. Winter, Sidney G. (2013): Habit, deliberation, and action: Strengthening the microfoundations of routines and capabilities. The Academy of Management Perspectives, Vol. 27, No. 2: 120–137. Winter, Sidney G. (2014): The future of evolutionary economics: Can we break out of the beachhead? Journal of Institutional Economics, Vol. 10, No. 4: 613–644. Winter, Sidney G. (2017): Pursuing the evolutionary agenda in economics and management research. Cambridge Journal of Economics, Vol. 2017, No. 41: 721–747. Winter, Sidney G., and Gabriel Szulanski (2001): Replication as strategy. Organization Science, Vol. 12, No. 6: 730–743. Wood, Wendy, Leona Tam, and Melissa Guerrero Witt (2005): Changing circumstances, disrupting habits. Journal of Personality and Social Psychology, Vol. 88, No. 6: 918–933. Wood, Wendy, and Dennis Rünger (2016): Psychology of habit. Annu. Rev. Psychol. Vol. 2016, No. 67: 11.1–11.26. Wood, Wendy, and David T. Neal (2009): The habitual consumer. Journal of Consumer Psychology, Vol. 19: 579–592. Zbaracki, Mark J., and Mark Bergen (2010): When truces collapse: A longitudinal study of priceadjustment routines. Organization Science, Vol. 21, No. 5: 955–972. Zollo, Maurizio, and Sidney G. Winter (2002): Deliberate learning and the evolution of dynamic capabilities. Organization Science, Vol. 13, No. 3, 339–351.

225

17 ORGANIZATIONAL ROUTINES Nathalie Lazaric

17.1

Introduction

Debate on organizational routines originated in the Carnegie School and Nelson and Winter’s (1982) book (Becker et al., 2005; Winter and Szulanski, 2001, Gupta et al., 2015, Baldessarelli, G. et al, 2022). The initial notion paid attention to organizational coordination, organizational docility, organizational knowledge, knowledge articulation and knowledge replication to build routines (Cohen, 1991; Cyert and March, 1963; Nelson and Winter, 1982; Simon, 1991; Winter and Szulanski, 2001). More recent work observes routines as collections of elements and considers parts of routines and organizational dynamics rather than routines as complete ‘entities’ (Feldman, 2000; Howard-Grenville and Rerup, 2017). This chapter aims to provide an overview of the discussion about organizational routines, diverse understandings of this notion with old and new debates. I investigate the historical foundations of and various debates on routines, to suggest a new lens through which to observe their organizational dynamics. The chapter is organized as follows. In Section 17.2, I draw on Herbert Simon’s (1955) insights to identify my point of departure for the study of routines. In this classical framing, cognition is central and organizational docility is a prerequisite for problem solving. In Section 17.3, I highlight the need to open this cognitive black box to observe the actions and patterns from routine dynamics perspective. I emphasize the notions of cognition and artefacts as opportunities for contributing to the routines debate. In Section 17.4, I argue that reflection and agency might be a promising starting point to understand the enactment of entrepreneurial action in case of radical uncertainty.

17.2

Story and foundations of organizational routines debate

Simon’s (1955) vision of organizational behavior has become classical framework for understanding the assumptions underlying cognition and problem solving. The idea of organizational docility is understood as important for observing routines and their foundations. However, recent research is providing new insights leading to reconsider cognition and organizational dynamics, and the epistemological foundations of routines.

226

DOI: 10.4324/9780429398971-19

Organizational routines

17.2.1

Cognition, problem solving, and routines

Simon’s framing focuses on the decision-making process, and how people make decisions in conditions of limited computational and informational resources (Simon, 1955, 1959; Brette et al., 2017). To provide a better understanding of the human mind, Simon collaborated with Allen Newell on some decisive and pioneering work in cognitive psychology and artificial intelligence. These developments in the behavioural foundations of organizations and exploration of the human mind resulted in important work on bounded rationality and a specific vision of problem solving. Simon’s view of the decision-making process underlines the implementation of cognitive mechanisms which to some extent are activated automatically. He goes on to develop an opposition between habit versus decision and coined the terms ‘routinized responses’ versus ‘problem-solving responses’ (March and Simon, 1958). Simon (1947, 1976): 88) argued that ‘habit permits the conservation of mental effort by withdrawing from the area of conscious thought those aspects of the situation that are repetitive [….] and permits attention to be devoted to the novel aspects of a situation requiring decision’. In this classical frame, rationality required a conscious choice among diverse options, and a choice process in which habits support rationality in the sense that they enable cognitive resources to be devoted to novel and complex situations. However, habits are ambiguous in that they both support rationality and are an obstacle to its development. Simon (1947, 1976): 90) believed that: In most cases, there seems to be a close relation […] between the spheres of attention and of rationality. That is, docility is largely limited by (1) the span of attention, and (2) the area within which skills and other appropriate behaviors have become habitual. Hence to a considerable extent, the limits upon rationality […] are resultants of the limits of the area of attention.

17.2.2

Docility a prerequisite for problem solving and organizational memory

The notion of docility in this process was important for Simon (1947) and refers to the situation where ‘[the individual] observes the consequences of his movements and adjusts them to achieve the desired purpose.’ Docility, then, is characterized by a stage of exploration and inquiry, followed by a stage of adaptation (Simon, 1947, 1976): 85). Simon (1947, 1976): 102) also underlined the social nature of mankind: human beings are members of social groups whether organizations or society, and this belongingness to a social order is not neutral for individuals. This vision of ‘framed’ social interactions and docility towards organizational goals, appears to be the solution to the limited cognitive resources of human beings. This epistemological foundation allowed Simon to focus on the decision-making process, and more specifically on the issue of problem solving (Newell and Simon, 1972). In this context: Habits and routines may not only serve their purposes effectively, but also conserve scarce and costly decision-making time and attention. For that reason, a very large part of an organization’s activities (or a person’s) is likely to proceed according to established rules and routines, which may be reviewed at shorter or longer intervals for possible revision. (Simon, 1947 [1997]: 89)

227

Nathalie Lazaric

Some dimensions of Simon’s legacy related to organizational and administrative processes are reflected in Nelson and Winter’s (1982) work where routines are the organizational memory allowing firms to accumulate knowledge and know how to achieve coordination and cognitive efficiency. This legacy was revived by the introduction of the Schumpeterian vision of innovation and entrepreneurship, and an evolutionary perspective on the original Carnegie School framing. The notion of routines was closely linked to the expansion and replication of knowledge from an entrepreneurial perspective. Cognition was both a survival mechanism and a behavioral regularity in a competitive environment but not really a process of knowledge creation by individuals (Winter, 1964).

17.2.3

Nelson and Winter’s observation of organizational routines

Nelson and Winter observed major military and civil technological programs and had close contacts with the Carnegie School. The unique experience obtained by Nelson and Winter gave them an understanding of the nature of the radical technical uncertainty inherent in all of creation. In the Carnegie School which was a veritable laboratory of ideas, March and Simon’s (1958) Organizations had a decisive impact on the whole scientific community (Nelson, 2006). Nelson and Winter, influenced by their reading of this work, began considering the dysfunctions of major military organizations and the role of institutions when launching technological innovations (Nelson and Winter, 1982). They applied their thinking to the firm which rather than responding to market signals, relied on specific rules. However, their initial Schumpeterian vision of the firm was moderated by their observation of these dysfunctions. Nelson and Winter rejected the ideas of routines evoked by Schumpeter and Simon and reformulated the notion of routines. These issues included where individual skills and the organization’s knowledge were located, and how they led to a successful production process. For Nelson and Winter (1982), the firm’s capacity for survival resides in its transformation induced either by a permanent reconfiguration of routines or a new combination of existing routines. Some transformations are internal to the firm since individual and collective knowledge is alive and not fixed entities. This dynamic is linked to the project’s Schumpeterian origins. Firms are continuously encouraged to innovate to survive, and innovation is the variable that induces transformations. This involves both internal and external transformations, whose origins may be difficult to trace (Winter, 2006). The notion of organizational routine which is considered a pillar of evolutionary thinking, is a source of both stability and change in the organization. The idea that changes to routines were more likely to be determined by changes in the environment (Cohen et al., 1996: 683) was widely held by adherents to the evolutionary theory of change (Anderson, 1994). However, Schumpeterian theory suggests two sources of renewal: the ‘combinatorics of routines’ which is based on the combination of sub-elements (Becker et al., 2006: 362), and the unreliable process of replication of internal elements. The firm’s ability to copy routines and/or to extend them is a source of competitive advantage. This may be an imperfect and costly mechanism, but it can also be a valuable source of evolution and change. Winter develops this idea in trying to explain the extension and renewal of knowledge bases in different competitive environments (Becker and Lazaric, 2003). Simon’s and Winter’s views are important but are not sufficient to explain how individuals act to renew and reconsider knowledge in their daily actions. These actions depend on the ideas of reflection and mindfulness and reconsider knowledge and cognition as 228

Organizational routines

something that is not innate but enacted and performed by actors (Lazaric and Denis, 2001, 2005; Levinthal and Rerup, 2006). This vision was reconsidered to some extent in Nelson and Winter (1982) but the main assumption that cognitive efficiency is essential for organizations to work smoothly persists. To open the black box of cognition, and to include mindfulness, individual voice, and endogenous change; a new framing called ‘routines dynamics’ emerged (Feldman et al., 2016), that will be discussed in Section 17.3.

17.3

Opening the black box and introducing action into the debate

Feldman and Pentland (2003: 96) define routines as ‘recognizable patterns of interdependent actions carried out by multiple actors’ and highlight that the main outcome of this is a ‘new understanding’ of routines which are constituted by and through actions. Making ‘action’ an essential building block of the micro-level dimension of routines allows study of the creation of new routines from an agency perspective, as emerging from the ‘relationship between specific actions and patterns of action’ (Pentland et al., 2012: 1485), and enables reconsideration of patterns of action and their cognitive roots.

17.3.1

Replication and emergence of new routines

When organizations are faced with ill-structured problems and unstructured decision processes, the building new routines may be complicated (Obstfeld, 2012). In such cases, the issue is not one of replicating the initial knowledge or resources but rather of identifying new patterns. Replication is complex and implies some degree of novelty for redesigning the old and adapting to the new context since knowledge cannot be replicated but must be discovered during this process (Becker et al., 2006). Knowledge replication and codification depend on a process of knowledge transfer and transformation which involves negotiation of its core and adaptation of its scope (Lazaric et al, 2003; D’Adderio, 2014). D’Adderio (2014) shows precisely why agency is critical and shows that goals may differ in the process leading to potential transformation of the template and the process of adjustment during knowledge replication. Creating new routines is an effortful task since it involves co-shaping of the ostensive and performative aspects of the routine. This duality represents a critical period of learning where the routines to be performed require the strong involvement of the actants through trial-and-error learning, experimentation, and improvisation to reduce uncertainty while articulating and trying to codify some of the knowledge. Rerup and Feldman (2011) highlight the combination of cognition and action for the emergence of new routines and the process of permanent adjustment that actors implement as they engage in trial-and-error learning. In short ‘[e]ach trial either replaced a specific performance in the (recruitment) routines with another specific performance or added a new performance to the existing set of performances’ (Rerup and Feldman, 2011: 603), showing it is not a fixed but rather a heterogeneous process which leaves space for creativity if complex problems arise. These empirical findings show that in the context of difficult problems ‘people are more like firefighters than they are like strategic decision makers’ (Rerup and Feldman, 2011: 604).

17.3.2

The role of artefacts as mediators between skills and routines

Artefacts are critical in this debate since they articulate knowledge and transform potential skills into future capabilities (Lazaric et al., 2003; D’Adderio, 2008; Cacciatori, 2012). It has 229

Nathalie Lazaric

been shown that artefacts are a source of memorization at the interface between the performative and ostensive aspects of routines (Cacciatori, 2008, 2012; D’Adderio 2008, 2011). There is a large amount of empirical evidence on the critical role of artefacts as mediators and manifestations of cognition. For instance, Orlikowski (1992) explains that human agency is mediated by manmade objects, while D’Adderio (2008) and Salvato (2009) discuss the criticality related to how agents create and make use of artefacts in their practice to maintain or change routines, and how artefacts serve to create external memories which are distributed among the actors. D’Adderio (2011: 197) highlights artefacts and acknowledges their importance as providers of ‘the glue that can hold action patterns together’. Cacciatori (2012) suggests that rather than being considered individually, artefacts should be seen as systems of objects involved in the creation of routines. She demonstrates that the emergence of new routines is mediated by the development of systems of artefacts able to reproduce problem solving structures at the heart of the routinization process. Artefacts also play a critical role in the building of standard operating procedures (Lazaric and Denis, 2005; D’Adderio, 2008). They mediate routines and skills and transform experience into potential new patterns (Cacciatori, 2012), paying attention to the micro-processes ‘through which patterns of actions are created and recreated from within’ (Dionysiou and Tsoukas, 2013: 184). The role of actors in designing artefacts and routines by inscribing their vision of ways of doing things is also important (Glaser, 2017). Artefacts encode knowledge and act as mediators between skills and routines (Cacciatori, 2012). They are rarely used in isolation; rather, they are used in systems which are important for stabilizing the firm structure. For instance, Lazaric and Denis (2005) show that handbooks describe ways of doing things, relying on sub groups of artefacts or subtasks such as the writing of procedures which convey and articulate knowledge at each step in the process. In short, systems of artefacts are arranged and are mutually reinforcing to stabilize existing performance and identify patterns that co-shape routines (Cacciatori, 2012). Some artefacts ‘contain a visual representation of knowledge. They include procedure, manual, reports, technical drawings and virtual prototypes’ (Cacciatori, 2012: 1362) and include representations of both the product and the process, that is, the way things are done and how and why they are done in a certain way, and why they make sense to the actors. These cognitive tools are entwined in the performative and ostensive dimension of routines (D’Adderio, 2011). Relatedly, Parmentier-Cajaiba et al. (2021) study biocontrol (pesticides without chemical inputs) and show that the process of permanent adaptation is critical for artefacts involved in stabilizing the process. Further, the use of artefact-as-coordination tool enables progress towards an ideal future situation (developing a new activity in line with the firm’s needs) and borrows from past situations (existing norms from the regulatory domain). Depending on the situation, the artefacts are viewed either as goals (ends-in-view) or as means (a spreadsheet, a database), showing that means and ends-in-view are not fixed but rather are ‘fluid and flexible’ (Dittrich and Seidl, 2018) because they are in the process of becoming (Tsoukas and Chia, 2002). I next discuss entrepreneurial actions and organizational dynamics.

17.4

The need of entrepreneurial actions and organizational dynamics in face of environmental degradation

The notion of agency which contrasts starkly with Carnegie School thinking, enables a new way of defining and naming routines with actors who are knowledgeable and often reflexive 230

Organizational routines

(Feldman et al., 2016). This provides a new vision of cognition for understanding the process and conditions allowing such reflexivity (Dittrich et al., 2016; Feldman, 2016). This allows a nice entry to understanding the human mind in action, and the limits of individual decision making, allowing a potential renewed source of inspiration for entrepreneurial actions.

17.4.1

New patterns to build and new challenges to handle: The case of environmental degradation

According to Dosi (1988), structures and scientific institutions contextually define the needs that are meant to be fulfilled, and the scientific principles and technology to be used in a given industry. Dosi (1988:1127) defines a technological paradigm ‘as an exemplar – an artifact that is to be developed and improved … and a set of heuristics’. Heuristics support problem solving and create shortcuts which allow individuals and organizations to operate in situations of uncertainty (Kahneman et al., 1982). Parmentier-Cajaiba et al.’s (2021) aforementioned study of a biocontrol start-up, show how much effort was involved in creating new routines, since the company had to translate all the biological practices related to the old paradigm (based on chemical principles) and negotiate their validity and existence. I argue here that we need to investigate this evolutionary process in more depth, including the creation of new path-dependent patterns of actions to understand how new ostensive patterns are accepted, validated, built, and re-built in a dynamic process. Pentland and Ju Jung (2016) provide significant insights which require consolidation to obtain a better understanding of the weight of new heuristics in patterning and the current heuristics performed by actors. Huge uncertainties such as those described in the grand challenges literature (Ferraro et al., 2015), require individuals and organizations to be ‘active experimenters’ and to identify some actions while solving new problems (for instance, in agricultural practice, how to reduce mildew on vines using non-chemical treatments). The need to find new patterns and to handle grand challenges among other issues may create significant internal and external tensions during experimentation.

17.4.2

Solving problems differently and building new routines through creative actions

Pentland (1995) proposes ‘grammars of action’ to describe routines where the grammars define the set of possibilities and variations related to a specific language, and the actions are the routines required to achieve the task (Pentland, 1995; Pentland and Reuter, 1994). Indeed, ‘an organizational routine is not a single pattern but rather a set of possible patterns’ (Pentland and Reuter, 1994: 491). Thus, solving problems, and ways of solving them open diverse opportunities to challenge (or not) habitual beliefs and patterns of interdependent actions. For instance, Mathews (2010) highlighted the significance of understanding how routines are created and how this process helps to provide a coherent, general account of strategic entrepreneurship. Mathews (2010: 224) defines entrepreneurship as ‘the activity that drives the economy in new directions through recombination of resources, activities and routines by firms. According to Mathews (2010), the effortful actions of entrepreneurs bring routines to life and set them in motion. Since ‘we still have much to learn about how process and context interact to shape the outcome of entrepreneurial efforts’ (Aldrich and Martinez, 2001: 41), the understanding mechanisms that enable entrepreneurial actions to directly shape 231

Nathalie Lazaric

organizational routines are critical. An entrepreneurial bricolage perspective suggests that individuals use scarce resources to create new organizational structures, solve new problems, and exploit new opportunities (Linna, 2013). Along these lines, Levi-Strauss (1966) proposed the concept of bricolage to suggest the creation of something new from the combination and transformation of existing resources (see Parmentier-Cajaiba et al., 2021, for an illustration). The combination of resources is at the core of the concept of bricolage. Individuals engaged in bricolage enact their environment and with other organizational actors co-shape ‘what is desired and feasible with the resources they have at hand’ (Janssen et al., 2018: 453). Bricolage is not just about limited resources; it is about transforming limited resources into useful outputs and may be a source of creativity through the recombination of existing resources in a context of grand challenges. Individuals and entrepreneurs may be more prone to be ‘active experimenters’ and may engage in different types of reflection for thinking about new patterns and embracing new values that shape their ostensive understandings of routines. In most cases, this is a long and uncertain process punctuated by unexpected problems. Thus, problem solving and ‘grand challenges’ could be reexamined based on Archer’s (2003, 2010) insights about different levels of reflection. Diverse types of ‘self-talk’ at the individual level could enable actors and entrepreneurs to rethink patterns of actions performed by. New forms of robust action and creativity are essential in the search for new ways of doing things. In this regard, Archer’s (2010) insights are worthy of reconsideration, especially in the context of creativity and new entrepreneurial actions in a context of environmental degradation.

17.5

Conclusion

Since the seminal work of Nelson and Winter (1982), much has already been achieved to understand routines their replication and the role of artefacts in this process. New insights provided by Feldman (2000) and Pentland (1994) integrate new elements of the observations for scrutinizing the internal dynamic of routines and a new understanding of the role of individuals in this process with the notion of agency. Debate introduced by routines dynamics provide a better understanding of the combination of sub-elements and replication of internal elements which might spawn entrepreneurial action. New patterns are built which show how cognition is rooted dynamically in the performance of action and is introduced in the action by the actors. Reflection is a critical element linking action to the patterns of actions. Archer and others offer some interesting notions for understanding the construct of intersubjective meaning and the role of the actors in the process of patterning (Dionysiou and Tsoukas, 2013). However, more work needs to be done to investigate the impact of the mutual constitution of the performative and ostensive aspects of routines and the impact of entrepreneurial action in this process.

References Aldrich, H.E., & Martinez, M.A. (2001). Many are called, but few are chosen: An evolutionary perspective for the study of entrepreneurship. Entrepreneurship Theory and Practice, 25, 41–56. Anderson, E. (1994). Evolutionary Economics: Post‐Schumpeterian Contributions. London: Pinter. Archer, M. (2003). Structure, Agency and the Internal Conversation. Cambridge: Cambridge University Press. Archer, M. (2010). Routines, reflexivity and realism. Sociological Theory, 28(3), 272–303.

232

Organizational routines Becker, M., & Lazaric, N. (2003). The influence of Knowledge in the replication of routines. Economie Appliquée, 16(3), 65–95. Becker, M.N., Lazaric, R.R., & Nelson, S.G. Winter. (2005). Applying routines in analysing organizations. Industrial and Corporate Change, 14(5), 775–791. Becker, M.C., Knudsen, T., & March, J.G. (2006). Schumpeter, Winter, and the sources of novelty. Industrial and Corporate Change, 15, 353–371. Brette, O., Lazaric, N., & de Vierera, V. (2017). Habit, decision making and rationality: Comparing Thorstein Veblen and Early Herbert Simon. Journal of Economic Issues, (3), 567–587. Cacciatori, E. (2008). Memory objects in project environments: Storing, retrieving and adapting learning in project-based firms. Research Policy, 37(9) 1591–1601. Cacciatori, E. (2012). Resolving conflict in problem-solving: Systems of artefacts in the development of new routines. Journal of Management Studies, 49, 1559–1585. Cohen, M.D., Burkhart, R., Dosi, G., Egidi, M., Marengo, L., Warglien, M., & Winter, S. (1996). Routines and other recurring action patterns of organizations: Contemporary research issues. Industrial and Corporate Change, 5(3), 653–697. Cohen, M.D. (1991). Individual learning and organizational routine: Emerging connections. Organization Science, 2(1), 135–139. Cyert, R.M., & March, J.G. (1963/1992). A Behavioral Theory of the Firm. Cambridge, MA: Blackwell Business. D’Adderio, L. (2011). Artifacts at the centre of routines: Performing the material turn in routines theory. Journal of Institutional Economy, 7, 197–230. D’Adderio, L. (2008) The performativity of routines: Theorising the influence of artefacts and distributed agencies on routines dynamics. Research Policy, 37, 769–789. D’Adderio, L. (2014). The replication dilemma unravelled: How organizations enact multiple goals in routine transfer. Organization Science, 25, 1325–1350. Dionysiou, D.D., & Tsoukas, H. (2013). Understanding the (re)creation of routines from within: A symbolic interactionalist perspective. Academy of Management Review, 38, 181–205. Dittrich, K., & Seidl, D. (2018). Emerging intentionality in routine dynamics: A pragmatist view. Academy of Management Journal, 61, 111–138. Dittrich, K., Guérard, S., & Seidl, D. (2016). Talking about routines: The role of reflective talk in routine change. Organization Science, 3(27), 678–697. Dosi, G. (1988). Sources, procedures, and microeconomic effects of innovation. Journal of Economic Literature, 26(3), 1120–1171. Feldman, M.S. (2000). Organizational routines as a source of continuous change. Organization Science, 11(6), 611–629. Feldman, M.S. (2016). Routines as process: Past, present, and future. In J. Howard-Grenville, C. Rerup, A. Langley, & H. Tsoukas, eds., Organizational Routines: How They are Created, Maintained, and Changed. Oxford, UK: Oxford University Press, pp. 23–46. Feldman, M.S. & Pentland, B.T. (2003). Reconceptualizing organizational routines as a source of flexibility and change. Administrative Science Quarterly, 48, 94–118. Feldman, M.S., Pentland, B.T., D’Adderio, L., & Lazaric, N. (2016). Beyond routines as things: Introduction of the special issue on routine dynamics. Organization Science, 27(3), 505–513. Ferraro, F., Etzion, D., & Gehman, J. (2015). Tackling grand challenges pragmatically: Robust action revisited. Organization Studies, 36(3), 363–390. Glaser, V.L. (2017). Design performances: How organizations inscribe artifacts to change routines. Academy of Management Journal, 60, 2126–2154. Gupta, A., Hoopes, D.G., & Knott, A.M. (2015). Redesigning routines for replication. Strategic Management Journal, 36, 851–871. Howard-Grenville, J.A., & Rerup, C. (2017). A process perspective on organizational routines”. In A. Langley, & H. Tsoukas eds, Sage Handbook of Process Organizational Studies. London, UK: Sage Publications, pp. 323–339. Janssen, F., Fayolle, A., & Wuilaume, A. (2018). Researching bricolage in social entrepreneurship. Entrepreneurship and Regional Development, 30 (3–4), 450–470. Kahneman, D., Slovic, P., & Tversky, A. (1982). Judgment Under Uncertainty: Heuristics and Biases. Cambridge: Cambridge University Press.

233

Nathalie Lazaric Lazaric, N., Mangolte, P-A, & Massué, M-L (2003). Articulation and codification of collective knowhow in the steel industry: Evidence from blast furnace control in France. Research Policy, 32, 1829–1847. Lazaric N., & Denis, B. (2001). How and why routines change: Some lessons from the articulation of knowledge with ISO 9002 implementation in the food industry. Economie et Sociétés, 585–612. Lazaric N., & Denis, B. (2005). Routinization and memorization of tasks in a workshop: The case of the introduction of ISO norms. Industrial and Corporate Change, 14, 873–896. Levi-Strauss, C. (1966). “The Savage Mind”. Chicago, US: University of Chicago Press. Levinthal, D.A., & Rerup, C. (2006). Crossing an apparent chasm: Bridging mindful and less mindful perspectives on organizational learning. Organization Science, 17, 502–513. Linna, P., (2013). Bricolage as a means of innovating in a resource-scarce environment: A study of innovator-entrepreneurs at the BOP. Journal of Developmental Entrepreneurship, 18 (3), 1350015. March, J. & Simon H. (1958). Organizations. NY: Wiley, 2nd ed., Oxford: Blackwell Publishers, 1993. Mathews, J.A. (2010). Lachmannian insights into strategic entrepreneurship: Resources, activities and routines in a disequilibrium world. Organizations Studies, 31, 219–244. Nelson R.R. (2006). Commentary on Sidney Winter’s “Toward a neo-Schumpeterian theory of the firm, Industrial and Corporate Change, 15(1), 145–149. Nelson, R.R., & Winter, S.G. (1982). An Evolutionary Theory of Economic Change. Cambridge, MA: Harvard University Press. Newell, A., & Simon H.A. (1972). Human Problem Solving. Englewood Cliffs, NJ: Prentice-Hall. Obstfeld, D. (2012). Creative projects: A less routine approach toward getting new things. Organization Science, 23(6), 1571–1592. Orlikowski, W.J. (1992). The duality of technology: Rethinking the concept of technology in organizations. Organization Science, 3, 398–427. Parmentier-Cajaiba, A., Lazaric, N., & Cajaiba-Santana, G. (2021). The effortful process of routines emergence: The interplay of entrepreneurial actions and artefacts. Journal of Evolutionary Economics, 31(1), 33–63. Pentland, B.T., & Reuter, H.H. (1994). Organizational routines as grammars of action. Administrative Science Quarterly, 39, 484–510. Pentland, B.T. (1995). Grammatical models of organizational processes. Organization science, 6(5) 541–556. Pentland, B.T., Feldman, M., Becker, M.C., & Liu, P. (2012). Dynamics of organizational routines: A generative model. Journal of Management Studies, 49(8), 1484–1508. Pentland, B.T., & Ju Jung, E. (2016). Evolutionary and Revolutionary Change in Path-Dependent Patterns of Action, in J. Howard-Grenville, C. Rerup, A. Langley, & H. Tsoukas eds., Organizational routines: how they are created, maintained, and changed. Oxford: Oxford University Press, pp. 96–113. Rerup, C., & Feldman, M.S. (2011). Routines as a source of change in organizational schema: The role of trial-and-error learning. Academy of Management Journal, 54(3), 577–610. Salvato, C. (2009). Capabilities unveiled: The role of ordinary activities in the evolution of product development processes. Organization Science, 20, 384–409. Simon, H.A. (1991). Models of My Life. NewYork, NY: Basic Books. Simon, H.A. [1947] (1976). Administrative Behavior. New York, NY: Free Press, 3rd edition. Simon, H.A. (1955). A behavioral model of rational choice. Quarterly Journal of Economics, 69(1), 99–118. Simon, H.A. (1959). Theories of decision-making in economics and behavioral science. American Economic Review, 49, 53–283. Tsoukas, H., & Chia, R. (2002). On organizational becoming: Rethinking organizational change. Organization Science, 13, 567–582. Winter S.G. & Szulanski, G. (2001). Replication as strategy. Organization Science, 12, 730–743. Winter, S.G. (1964). Economic “Natural Selection” and the Theory of the Firm. University of Michigan: Institute of Public Policy Studies, 4, pp. 225–272. Winter, S. (2006). Toward a neo-Schumpeterian theory of the firm, Industrial and Corporate Change. Oxford University Press, 15(1), 125–141.

234

18 MEMES Michael P. Schlaile, Walter Veit, and Maarten Boudry

18.1

Introduction

With this chapter, we aim to (re-)introduce the notion of memes into economic theory. For some, this endeavor may seem like flogging a dead horse; for others, a long overdue project of building bridges between different disciplines and fragmented approaches. Are memes nothing but a misleading metaphor for non-existent entities, wrongly alleged to be analogous to genes? Not so, we shall argue. The idea of memes as the units of cultural evolution has been around for almost half a century now, although most contemporary researchers in cultural evolution prefer to call them “cultural variants” or “cultural traits” (e.g., Schurz, 2021; Wilson, 1998). In a similar manner, evolutionary economists have proposed various candidate units in an economic context, including habits, ideas, modules, routines, rules, and utopias (e.g., Almudi et al., 2017a,b; Beinhocker, 2006; Breslin, 2016; Dopfer et al., 2004; Hodgson & Knudsen, 2010; Markey-Towler, 2019; Nelson & Winter, 1982). Considering the different intellectual histories of these concepts, it is unsurprising that evolutionary economics is a rather fragmented field (e.g., Hodgson & Lamberg, 2018; Witt, 2014). While memes have frequently faced criticism from many of these schools of thought (e.g., Roy, 2017; Chap. 6 in Hodgson & Knudsen, 2010), some have argued that they can serve as a common language for linking several of these concepts and approaches (e.g., Schlaile, 2021). Our chapter aims to shed light on this promise, while remaining cautious about overly ambitious claims to the effect that selfish memes can essentially explain all of human culture, a position Boudry and Hofhuis (2018) have criticized as panmemetics.1 Given the limited space and scope of this chapter, our contribution should be treated as an invitation for further work rather than a comprehensive presentation of a fully developed theory. Readers unfamiliar with the concept are referred to the excellent introductory article by von Bülow (2013), Dennett’s extensive work on memes (1995, 2006, 2017), and Chap. 2 in Schlaile (2021). The three main points we want to make here is that (i) evolutionary economists have been biased towards mostly intentional and “adaptive” processes of innovation and technological and economic change, neglecting unintentional and “maladaptive” evolutionary processes, (ii) the meme’s eye view (as opposed to an agent-centered view) still offers a

DOI: 10.4324/9780429398971-20

235

Michael P. Schlaile et al.

valuable perspective for evolutionary economics, and (iii) memes should best be regarded as units of informational structures—often containing instructions—that can be socially transmitted and recombined, thus affording the emergence of innovations. The chapter is organized as follows. First, we revisit and summarize important arguments for taking cultural evolution and the meme’s eye view more seriously. The subsequent section highlights the merits of viewing memes as informational entities that often include an element of instruction, thus providing a link to the rule-based approach to evolutionary economics. Next, we dismiss an overly reductionist view of memes as discrete and “independent” cultural elements, by viewing memes as embedded within complex systems. We then briefly turn to the memetics of creativity and innovation before we summarize our arguments in terms of the “five i’s of economemetics” and conclude our chapter with propositions for future interdisciplinary inquiries.

18.2

Cultural evolution, imitation, and the meme’s eye view

For the purpose of this chapter, we adopt the liberal definition of culture proposed by Boyd and Richerson: “Culture is information capable of affecting individuals’ behavior that they acquire from other members of their species by teaching, imitation, and other forms of social transmission” (Boyd & Richerson, 2005, p. 6, emphasis removed). There is ample literature on how particular cultural values and worldviews (including religious ideologies and practices) have influenced the emergence, success, and continued existence of economic systems and practices such as capitalism (e.g., Henrich, 2020; Hodgson, 2015; Weber, 1930, 2001; Schramm, 2008). However, contemporary evolutionary economists have focused mostly on the technological aspects of innovation and industrial change. By contrast, they have paid relatively little attention to the evolution of cultural value systems and belief systems and how they interrelate with technological and economic change.2 Both cultural and economic systems have been argued to evolve analogously to processes known from biological evolution (e.g., Dennett, 2017; Hodgson & Knudsen, 2010; Lewens, 2020; Veit, 2019a; Wilson & Gowdy, 2013; see also the discussions in Gagliardi & Gindis, 2019; Wilson & Kirman, 2016; Witt & Chai, 2019). In fact, as Ginsburg and Jablonka stress with reference to Charles Darwin’s selection theory: “The generality of the idea [of evolution by natural selection] allows it to be applied to disciplines as different as cosmology, economics, culture, and ethics, as well as to processes occurring in the brain” (Ginsburg & Jablonka, 2019, p. 65). This is not to say, however, that evolutionary processes across all systems involve the exact same “mechanism”, since some cultural evolutionary processes are more “Darwinian” than others (Dennett, 2017), for instance by being more or less gradual or more or less goaldirected (Dennett, 2021; Mesoudi, 2021). The next important step is to acknowledge that cultural evolution involves, at least in part, the replication of units of information. In the social environment we live in, information is largely socially distributed not only across different media but also across different minds. To make use of this, humans have become masters of imitation and learning, information sponges that absorb all sorts of information from our social environments. The importance of cultural replication and imitation has also been affirmed by researchers studying adaptive behavior and cognition,3 who have identified several imitation heuristics (e.g., Boyd & Richerson, 2005; see also Chap. 8.3 in Godfrey-Smith, 2009).4 A common definition of a meme is an “element of a culture or system of behaviour passed from one individual to another by imitation or other non-genetic means” (Oxford Dictionaries, undated). Though this succinct definition captures the essence of the concept, 236

Memes

it leaves open its ontological status. What sort of thing is a meme exactly, and where should we locate it? This leads us to the first important way to classify memetic approaches. On the one hand, there are approaches that seek to identify memes with some material substrates, such as brain structures, artifacts, or behaviors. A different approach regards memes as abstract and (substrate-neutral) informational entities. We could call this first distinction “material” vs. “informational” approaches. Meme theorists also differ with respect to how much of human culture they see as “viral” or “parasitical”, and how exactly they define those terms (see also Blute, 2010). For some theorists, all of human culture should be regarded as swarms of viral memes that infect human brains with purposes and interests of their own (e.g., Blackmore, 2000; Stanovich, 2005, for details). Other meme theorists see more room for human intentionality and design and restrict the concept of “viral” memes to certain deleterious cultural beliefs and practices. In order to understand the image of viral or parasitical culture, we have to adopt what meme theorists call the meme’s eye view (Dawkins, 1993; Dennett, 1995, 2006). The best way to understand this key concept is to contrast it with its alternatives. In traditional accounts of culture—and even in most evolutionary approaches to culture—it is taken for granted that cultural ideas and artifacts serve some useful function or provide some benefit to human beings (for a critique, see Hofhuis, 2022; Edgerton, 1992). Or more precisely, to the extent that they have some function, we human beings must be the beneficiaries. It is human beings, after all, who select, discard, or retain cultural ideas and artifacts. Who else could benefit? By contrast, the meme’s eye view invites us to adopt the perspective of the cultural items themselves. Because cultural items (memes) replicate and form chains of transmission, cultural evolution will select the memes that are most successful at dissemination. This sets up an evolutionary dynamic that is relatively autonomous from human agents and may produce forms of cultural design whose functional rationale is opaque to them. In some cases, the “interests” of memes and their human carriers (or “hosts”) will align pretty well: we select and spread some memes because we find them appealing, and they enhance their own propagation by appealing to us. But in the most interesting cases, the interests of memes and their carriers diverge: “parasitical” memes spread because they are contagious and catchy, despite the fact that they are harmful to their human carriers. For instance, conspiracy theories are prime examples of highly attractive and contagious memeplexes because their internal structure renders them self-validating: once you adopt the idea of a grand conspiracy theory, every form of adverse evidence can be turned around and presented as positive evidence (Boudry, 2020, 2022; Law, 2011). Despite these attractive features, conspiracy theories wreak a lot of havoc in society. Other examples of parasitical memes include superstitions, pseudoscience, addictions, bad habits, and ear worms (e.g., Dennett, 2017; Boudry & Hofhuis, 2018). To understand the functional rationale of such viral or parasitical forms of culture, we have to adopt the meme’s eye view. By doing so, memeticists draw out patterns of human culture that are invisible if we only consider the interests of human agents (e.g., Boudry, 2018a; Boudry & Hofhuis, 2018; Hofhuis, 2022). Note that this discussion also links to a more general debate on functionalism in institutional theory (e.g., Chap. 5 in Krul, 2018): Are institutions (such as prevalent rules, norms, laws, and regulation quite generally) always intentionally and consciously established for the benefit of society by (more or less) rational agents, or are they rather the result of often unintentional and historical/path-dependent cultural evolutionary processes (see also Rosenberg, 2021; Runciman, 2015, on a related discussion)? In the latter case, they may also lead to a lock-in of unsustainable and destructive practices and socio-technical regimes 237

Michael P. Schlaile et al.

(e.g., Geels, 2002; Edgerton, 1992). In the same vein, as the recent literature on responsible innovation highlights, technological innovation, which is arguably a specific type or embodiment/manifestation of cultural evolution (e.g., Richerson & Christiansen, 2013), does not necessarily imply “progress” (e.g., Blok & Lemmens, 2015; Ruse, 1993; Schlaile et al., 2017, 2018).

18.3 Memes as information and instruction In line with Boyd and Richerson’s definition of culture adopted above, we adopt an informational approach to memes, which is not committed to any particular physical substrate and is therefore better suited for bridging (seemingly) conflicting approaches across disciplines. Following “informationalists” such as Boudry (2018b) and Dennett (2006, 2017), we believe that memes are most usefully thought of as pieces of abstract information, which can be instantiated in different media. In our view, the informational perspective defuses many of the most common objections against memes. In particular, many theorists have opposed the concept of memes because they claim that, unlike in biological evolution, there is no physical structure that can be identified as the unit of replication (see also Roy, 2017). In other words, there is no physical analogue to the gene in the cultural realm. To talk of memes, according to critics, is to admit a phlogiston-like entity in cultural evolution. Tying the success of cultural evolution to finding the cultural analogue of genes, they fear, is a theoretical dead-end. But to take the gene/meme analogy literally is to misunderstand the role of analogies and metaphors in the sciences. In the history of science, metaphors and analogies have often enabled important breakthroughs despite being treated with suspicion by philosophers of science, especially ones which map concepts across distant domains. Even though no analogyis perfect, they help us to extend the reach of our mind and see connections and relations that were previously invisible (Veit & Ney, 2021; Boudry, Vlerick & Edis, 2022). Firstly, this opposition to the concept of memes rarely recognizes that the concept of genes itself is far from straightforward. As Wilkins and Bourrat (2022) put it, in many critiques of cultural replication “[a]n overly idealized view of Mendelian genetics is contrasted to a much more realistic view of cultural change”. Various definitions of the gene across the biological sciences appear to be irreconcilable. Pluralism rules. It is true that genes appear more localizable and easier to pinpoint than memes, being associated with a single type of molecule (DNA or RNA), but this is not essential to the notion of a gene. From an evolutionary perspective, the most useful definition of a gene is as an abstract piece of information, not as a particular molecule. Genes, as Williams (1992) and others have pointed out, should not be identified with DNA but with the information carried by DNA. A gene is a piece of abstract information that is relatively stable and can be tracked across generations. It is, as Williams put it, “that which segregates and recombines with appreciable frequency” (Williams, 1966/2019, p. 24), regardless of whether the information is spread across the genome or unified and isolated. Unlike physical definitions of genes, this informational definition can be easily extended to the cultural realm (see also Ball, 1984). A meme of a music tune or an idea is a piece of abstract information that “segregates and recombines with appreciable frequency” in human cultures. It can be stored in a human brain, a digital mp3 file, or on a piece of sheet music. It can be written into a diary or recorded in the form of sound waves. Memes and genes on this view are not mere parallels, they are essentially the same type of abstract entity. To those who reject the meme concept 238

Memes

because it cannot be physically identified—or because it smacks of dualism—it must be asked whether they also deny the informational gene concept defended by Williams and others. Information can be stored in all kinds of different ways. In the case of biology, the carriers are usually DNA or RNA, but this is merely incidental. A digital computer file describing the genetic sequence of, say, the severe acute respiratory syndrome coronavirus type 2 (SARS-CoV-2), contains the same information as the RNA molecules inside the virus itself, and the information can be transcribed from one medium to another. In the cultural domain, there is a much wider variety of different media, but the evolutionary dynamic is exactly the same. Memetics, and in particular the meme’s eye view, makes sense of the dynamics of information transfer in the cultural world. How can information move from one physical instantiation to another—whether this is neural, language, pictures, or anything else for that matter—and how does this information evolve? A similar reply can be given to the objection that cultural evolution does not involve simple and straightforward “replication” like in the case of genes, but rather heavily relies on reconstruction (Sperber, 1996, 2000; Hodgson & Knudsen, 2010). This perceived contrast with gene replication, too, underestimates the messiness of biological reality. The genomes of two cells resulting from mitosis are not exact replicas, since they differ in numerous ways (they are wound up and folded differently, and their lower-level molecular structure differs in countless ways). They are only “replicas” of each other to the extent both can be regarded—at the right level of abstraction—as embodying a certain amount of information, and because their differences will be normalized and ignored when they are transcribed and read by ribosomes (Boudry, 2018b). It is also important to note that in both biological and cultural evolution, replicators have frequently been seen as containing instructions (see also Cloak, 1975). Dennett, for instance, argues that memes “are ‘prescriptions’ for ways of doing things” (Dennett, 2017, p. 211). Similarly, Heylighen and Chielens (2009) have likened memes to production rules (IF condition, THEN action), and this sentiment is prominently captured by Ostrom’s statement that “rules are sets of instructions for creating an action situation … As such, rules are broadly analogous to genes, which are sets of instructions for creating a phenotype. Rules are memes rather than genes, but it is helpful to think about some of the similarities between genes and memes” (Ostrom, 2006, p. 116). This brings us to an important connection between memes and the rule-based approach (RBA) to evolutionary economics, championed by Dopfer et al. (2004) and Dopfer and Potts (2008, 2019). For the sake of brevity, we cannot go into much detail here, but it should be acknowledged that both memetics and the RBA could gain from more integration. For instance, the elaborate rule taxonomy developed by Dopfer and colleagues, which differentiates between various subject and object rules along an evolutionary micro-meso-macro trajectory (Dopfer et al., 2004, Dopfer & Potts, 2008, 2019), provides an analytical schema that can also help memeticists to focus their attention on the instructional part of a cultural information present at multiple levels ranging from individuals to firms to whole economic systems. In turn, the RBA may profit from taking up some of the analytical instruments available in contemporary meme theory and recent propositions to operationalize memes (e.g., Schlaile, 2021, esp. Chap. 3).

18.4

No meme is an island: Why interconnection is key

Memes exist in complex interrelationships with other memes and their “environment”. In fact, as Dennett (1995, p. 144) puts it, “no meme is an island”, since memes may both 239

Michael P. Schlaile et al.

promote or impede the variation, selection, and retention of other replicators (genes and memes alike) (see also von Bülow, 2019, on a related note). In the same vein, Weeks and Galunic stress: “We cannot look at memes in isolation. When conceptualizing how culture evolves through a process of the variation, selection, and retention of memes, we must explicitly take into account the fact that memes only make sense when we look at their patterns of combination” (Weeks & Galunic, 2003, p. 1317). But what does that mean, exactly? By drawing on Hodgson’s (2011) notion of a complex population system5 in combination with an informational approach to memes (as described above) and Simon’s (1971) well-known statement that the overabundance of information leads to a scarcity of attention, which thus needs to be focused accordingly, memes can be regarded as “competing” for the “scarce resource” of attention. More precisely, the extent to which memes draw our attention depends not only on how attractive their own informational content is but also on how compatible they are with other information sources, especially other memes in the system (Schlaile, 2021, Chap. 3). These compatibility relations can be depicted as links of a meme network. Despite the fact that economics studies complex systems, economists—especially in the dominant traditions—have been rather reluctant to take up approaches from complexity science, unlike other sciences of complex systems such as ecology, climate science, and evolutionary biology. There has been a temptation in economics to rely on as few models as possible. Much of the opposition to memes in economics, we fear, rests on the idea that “less is better”.6 This, we think, is a mistake. What is needed is a recognition that science requires what Veit (2019b, p. 93) calls “model pluralism”, that is, the idea that “for almost any aspect x of phenomenon y, scientists require multiple models to achieve scientific goal z” (see also Veit, 2021, 2023). What those interested in memes are studying includes the informational aspect as well as the unintentional, potentially even harmful effects of cultural change. These aspects of the economic system are rarely studied explicitly. By applying the network representation of complex systems, we can observe interconnections at the level of the memes themselves (a network of, e.g., knowledge units embodied in the mental representations of economic agents) as well as the more frequently analyzed social and economic networks of the agents within, say, an innovation system (e.g., Schlaile, 2021, esp. Chap. 5). This interconnectedness of different levels of complex systems present within an economy also links back to the literature on cultural multilevel selection: While we acknowledge that selection processes in an economy can occur at multiple levels (e.g., Field, 2008; Waring et al., 2015), we would also argue that most literature on cultural multilevel selection does not pay much attention to network complexity at the “lower” meme level. In other words, while multilevel selection theory aptly captures the tensions between self-interested and more prosocial behaviors of people (e.g., Atkins et al., 2019), the interconnections among the informational instructions (i.e., memes) embodied within those people are usually not addressed. We thus side with Velikovsky (2016, 2018) in highlighting the nested hierarchy or “holarchy” (Koestler, 1967) of selection processes. In our framework, memes are “holons” or fractal entities that belong to larger memeplexes, which are in turn part of complex systems more generally (e.g., Schlaile, 2021, Chap. 3, for details). This is in line with Koestler’s argument that “‘wholes’ and ‘parts’ in … [an] absolute sense just do not exist anywhere, either in the domain of living organisms or of social organisations. What we find are intermediary structures on a series of levels” (Koestler, 1967, p. 48). 240

Memes

18.5

Is everything a remix? Creativity and innovation from a memetic perspective

One of the most remarkable features of the human mind and our behavioral repertoire is our almost unlimited range of options. We can combine and transform ideas and copy them from others. Indeed, the processes of copying, transforming, and (re-)combining, often summarized under the umbrella of “remix” (Ferguson, 2015), exhibit striking overlaps with Darwinian evolutionary processes, especially variation, selection, recombination, retention, and transmission (Schlaile, 2021, Sect. 7.2). Evolution often results in the increasing creativity of actors—in the sense of their being able to extract information from the environment in new and useful ways in order to respond to their Umwelt (Veit, 2022). Memes are the units of this information. Memetic creativity can thus be understood as the degree of a human carrier’s “susceptibility” to taking up and recombining memes in novel ways that may help the carriers to learn and flexibly respond in complex social environments, opening a space for innovation and new ideas that can potentially benefit us and those around us (similar to how evolvability helps species to react to changing environments). Or, to use Kauffman’s (2000) terms, evolution (both biological and cultural) is about reaching the “adjacent possible” time and again, thus accumulating creative changes in complex and path-dependent ways (Johnson, 2010; Ridley, 2020). The meme’s eye view makes these processes less mysterious, putting creativity firmly within a naturalist view of the mind. Yet, some feel unease about this view of how the mind operates (e.g., see also Kronfeldner, 2011; Mesoudi, 2021; Simonton, 2003; Wagner, 2019, on related discussions). Are we merely the breeding ground for ideas (memes) we have picked up somewhere before? Interestingly, in line with the memetic approach to creativity (see also Sect. 7.2 in Schlaile, 2021), Tarde already maintained at the beginning of the 20th century that “every invention and every discovery consists in the interference in somebody’s mind of certain old pieces of information that have generally been handed down by others” (Tarde, 1903, p. 382). Is creative genius mere plagiarism, as somewhat jokingly mentioned by Ball (1984)? In the public imagination, genius and creativity are frequently conceived as inexplicable outbursts of imagination, as if new ideas come down from heaven like a lightning strike. In the same vein, innovation economists have long criticized the neoclassical economists for treating knowledge as an intangible good with some of the features of a public good. In this view, knowledge flows freely between actors or appears to fall “like manna from heaven”, a point that Robert Solow is frequently credited for pointing out (see also Urmetzer et al., 2018, for references and further discussions on this issue). But our minds are not blank slates and are always already teeming with memes. We make do with what we have. And since we are unlike any other animal (though some smart animals like octopuses and corvids engage in similar activities), we are able to absorb all kinds of information from our environment, mixing it into novel ideas and behavioral innovations (see also Dennett, 2021). As innovation economists have long acknowledged, innovations are often the emergent outcomes of interactions among various different actors weaving complex networks of cooperation, competition, and other forms of interdependence, frequently captured by notions like innovation networks, innovation systems, and innovation ecosystems (e.g., Breslin et al., 2021; Buchmann & Pyka, 2012; Rakas & Hain, 2019). In fact, innovation is often not the work of foresight genius or top-down oversight, but of unplanned trial and error, incremental steps, and endless recombination (Ridley, 2020). A historical and evolutionary approach to innovation takes some of the apparent genius away, or rather

241

Michael P. Schlaile et al.

distributes it over many different agents. In cultural and economic evolution, just as much as in biological evolution, Leslie Orgel’s second rule applies: evolution is cleverer than you are. As Potts (2019) has recently argued, this evolutionary, uncertain, and collective nature of innovation makes it a collective action problem, namely of pooling knowledge and resources, establishing institutions for cooperation, and deciding which memes in the sense of knowledge units should be combined. In this regard, a memetic approach to creativity can also provide new impetus to recent discussions on innovation policy and intellectual property rights, and potentially lend a naturalistic support to approaches like open innovation (Chesbrough, 2003) or free innovation (von Hippel, 2017), though in the latter regard by focusing on the meme level of analysis instead of focusing mainly on the human actors. It should go without saying that we do not intend to abolish intellectual property rights or recommend allowing other companies to simply copy an existing product (or process or service, etc.). Rather, we propose to facilitate the selection of an institutional framework within an innovation system that does not unnecessarily impede the merger of memes/ knowledge among companies.

18.6 Summary and conclusion The combination or synergy of memetics and (evolutionary) economics has been called economemetics (Schlaile, 2021). This neologism should not be misunderstood as a new discipline but rather as a perspective that aims at consilience and bridging fragmented approaches. In this regard, the key take home messages from the above discussion can be summarized with the five i’s of economemetics: Memes can be understood as units of information that often contain rule-based elements of instruction, which may be transmitted via imitation and other processes of communication and social learning. Moreover, variation, selection, and retention (and “remix”) of memes lead to innovations that emerge from the interconnections of both memes and economic agents in complex (often multi-level) networks. Importantly, compared to other evolutionary approaches, memetics is distinctive for adopting the meme’s eye view, which considers the “interests” of cultural elements themselves. Memes can be useful or beneficial to human agents, but they can also be “parasitical” cultural elements that further their own propagation despite harming their human hosts. With respect to economics, the meme’s eye view complements existing approaches, for example, in innovation economics by naturalizing creativity and innovation. Rather than resulting from strokes of genius or virtually falling down from the sky, cultural innovation usually involves many rounds of variation, selection, and recombination within complex networks of cooperating and competing individuals and organizations. In this sense, (econo-)memetics makes creativity less “mysterious” but also less individualistic, bringing it down to earth again. There are multiple pathways to pursue in future research, including theoretical clarifications on the nature of “information” and further exploration of the potential synergies between memetics and the RBA mentioned near the end of the section on “Memes as information and instruction”. Moreover, some striking overlaps seem to exist not only with concepts developed in evolutionary economics (i.e., habits, ideas, routines, rules, etc.) but also with notions like frames and narratives (e.g., Riedy & Waddock, 2022) and findings from adjacent fields such as (bio)semiotics (e.g., Fomin, 2019; Herrmann-Pillath, 2021) that should 242

Memes

be taken up in future conceptual and empirical research (e.g., Schlaile, 2021, Chap. 8). By focusing on memes as the evolutionary foundations (or “building blocks”) of worldviews and belief systems, we may even shed new light on the complex dynamics of “normative dimensions” of economic systems (e.g., Schlaile et al., 2017) and the resulting paradigms that could block or promote transitions towards more sustainable modes of production and consumption (Schlaile et al., 2022). Finally, model pluralism also gives rise to different ways of operationalization. More precisely, empirical studies on memes can resort to a wide variety of tools and methods even beyond those known from evolutionary biology and anthropology, including but not limited to text mining approaches (e.g., sentiment analysis, topic modeling, etc.) that so far had little impact in economics. In conclusion, we think the time has come for a renewed and interdisciplinary engagement with memes in economics.

Notes 1 Our chapter builds on and extends earlier arguments, some of which have been previously published independently by the authors of this chapter, for example, in Schlaile (2021) and Boudry (2018a, b). This work has not received any particular financial support, but Michael Schlaile gratefully acknowledges partial funding through the German Academic Exchange Service (DAAD) (Project ID: 57563063), VolkswagenStiftung (Project ID: 99 116), and the Federal Ministry of Education and Research (grant number 031B0751), Maarten Boudry received support from the Research Foundation Flanders, and Walter Veit has received funding as part of a project from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement number 101018533). 2 Of course, this should not imply that cultural change or even cultural evolution with an explicit Darwinian connotation has received no attention from evolutionary economists (e.g., see Herrmann-Pillath, 2010, 2013, 2021), as also several scholars at the intersection of economic history, institutional economics, and evolutionary economics have shown (for recent examples, see Hodgson, 2019, or contributions in Gagliardi & Gindis, 2019; Witt & Chai, 2019). However, it is fair to say that technological change, the creation and diffusion of economically useful knowledge and innovations, and the dynamics of sectors, industries, and various types of innovation systems have received much more attention from evolutionary economists than the evolutionary dynamics of value(s) and belief systems. 3 Note, however, that we are not committed to the existence of any straightforward mechanism of high-fidelity replication, regardless of how that is construed ( Charbonneau 2020). To a large extent, cultural transmission events involve a complicated process of reconstruction rather than a straightforward process of copying. If we want to study the evolution of memes or cultural variants on a population level, however, we can abstract from those lower-level complications ( Boudry, 2018b; Acerbi 2019). No matter how the process of cultural transmission is achieved, in the aggregate it often results in cultural traditions that are remarkably stable and persistent, and thus “faithfully” preserved. See also von Bülow (2013, 2019) on related discussions. 4 Moreover, the French sociologist Gabriel Tarde should be mentioned as an important figure in imitation research ( Blute, 2022) as he has been considered a “forefather” of memetics (e.g., Marsden, 2000; Schmid, 2004), of elements of Schumpeter’s works (e.g., Barry & Thrift, 2007; Kobayashi, 2015; Taymans, 1950), and of diffusion research ( Katz, 2006; Kinnunen, 1996; Rogers, 2003). 5 According to Hodgson (2011, p. 309), “complex population systems contain multiple varied (intentional or non-intentional) entities that interact with the environment and each other. They face immediately scarce resources and struggle to survive, whether through conflict or cooperation. … They adapt and may pass on information to others, through replication or imitation.” 6 Note that this is in line with the ongoing critiques by heterodox economists and initiatives such as Rethinking Economics ( https://www.rethinkeconomics.org/), the German Network for Pluralist Economics ( https://www.plurale-oekonomik.de/), and others.

243

Michael P. Schlaile et al.

References Acerbi, A. (2019). Cultural evolution in the digital age. Oxford: Oxford University Press. Almudi, I., Fatas-Villafranca, F., Izquierdo, L. R., & Potts, J. (2017a). The economics of utopia: A coevolutionary model of ideas, citizenship and socio-political change. Journal of Evolutionary Economics, 27, 629–662. DOI: 10.1007/s00191-017-0507-7. Almudi, I., Fatas-Villafranca, F., & Potts, J. (2017b). Utopia competition: A new approach to the micro-foundations of sustainability transitions. Journal of Bioeconomics, 19, 165–185. DOI: 10. 1007/s10818-016-9239-2. Atkins, P. W. B., Wilson, D. S., & Hayes, S. C. (2019). Prosocial: Using evolutionary science to build productive, equitable, and collaborative groups. Oakland, CA: Context Press. Ball, J. A. (1984). Memes as replicators. Ethology and Sociobiology, 5, 145–161. DOI: 10.1016/01623095(84)90020-7. Barry, A., & Thrift, N. (2007). Gabriel Tarde: Imitation, invention and economy. Economy and Society, 36, 509–525. DOI: 10.1080/03085140701589497. Beinhocker, E. D. (2006). The origin of wealth: Evolution, complexity, and the radical remaking of economics. Boston, MA: Harvard Business School Press. Blackmore, S. (2000). The meme’s eye view. In R. Aunger (Ed.), Darwinizing culture: The status of memetics as a science (pp. 25–42). Oxford: Oxford University Press. Breslin, D. (2016). What evolves in organizational co-evolution? Journal of Management & Governance, 20, 45–67. DOI: 10.1007/s10997-014-9302-0. Blok, V., & Lemmens, P. (2015). The emerging concept of responsible innovation. Three reasons why it is questionable and calls for a radical transformation of the concept of innovation. In B.-J. Koops, I. Oosterlaken, H. Romijn, T. Swierstra, & J. van den Hoven (Eds.), Responsible innovation 2: Concepts, approaches, and applications (pp. 19–35). Cham: Springer. Blute, M. (2010). Darwinian sociocultural evolution: Solutions to dilemmas in cultural and social theory. Cambridge: Cambridge University Press. Blute, M. (2022). Gabriel Tarde and cultural evolution: The consequence of neglecting our Mendel. Journal of Classical Sociology. DOI: 10.1177/1468795x221136830. Boudry, M. (2018a). Invasion of the mind snatchers. On memes and cultural parasites. Teorema, 37, 111–124. Boudry, M. (2018b). Replicate after reading: On the extraction and evocation of cultural information. Biology & Philosophy, 33, 27. DOI: 10.1007/s10539-018-9637-z. Boudry, M., & Hofhuis, S. (2018). Parasites of the mind. Why cultural theorists need the meme’s eye view. Cognitive Systems Research, 52, 155–167. DOI: 10.1016/j.cogsys.2018.06.010. Boudry, M. (2020). The warped epistemology of conspiracy theories, Blog of the American Philosophical Association. Retrieved from: https://blog.apaonline.org/2020/09/08/the-warpedepistemology-of-conspiracy-theories/. Boudry, M. (2022). Why we should be suspicious of conspiracy theories: A novel demarcation problem, Episteme. DOI: 10.1017/epi.2022.34. Boudry, M., Vlerick, M., & Edis, T. (2022). Demystifying mysteries. How metaphors and analogies extend the reach of the human mind. In S. Wuppuluri, & A. C. Grayling (Eds.), Metaphors and analogies in sciences and humanities: Words and worlds (pp. 65–83). Cham: Springer. Boyd, R., & Richerson, P. J. (2005). The origin and evolution of cultures. Oxford: Oxford University Press. Breslin, D., Kask, J., Schlaile, M., & Abatecola, G. (2021). Developing a coevolutionary account of innovation ecosystems. Industrial Marketing Management, 98, 59–68. DOI: 10.1016/j.indmarman. 2021.07.016. Buchmann, T., & Pyka, A. (2012). Innovation networks. In M. Dietrich, & J. Krafft (Eds.), Handbook on the economics and theory of the firm (pp. 466–482). Cheltenham: Edward Elgar. Charbonneau, M. (2020). Understanding cultural fidelity. The British Journal for the Philosophy of Science, 71, 1209–1233. DOI: 10.1093/bjps/axy052. Chesbrough, H. (2003). Open innovation: The new imperative for creating and profiting from technology. Boston: Harvard Business School Press. Cloak, F. T. (1975). Is a cultural ethology possible? Human Ecology, 3, 161–182. DOI: 10.1007/ BF01531639.

244

Memes Dawkins, R. (1993). Viruses of the mind. In B. Dahlbom (Ed.), Dennett and his critics: Demystifying mind (pp. 13–27). Oxford: Blackwell. Dennett, D. C. (1995). Darwin’s dangerous idea: Evolution and the meanings of life. London: Simon & Schuster. Dennett, D. C. (2006). Breaking the spell: Religion as a natural phenomenon. New York: Penguin. Dennett, D. C. (2017). From bacteria to Bach and back: The evolution of minds. New York: W. W. Norton. Dennett, D. C. (2021). Foreword to special issue: Cultural evolution and generalized Darwinism: Theory and applications. American Philosophical Quarterly, 58, 1–5. DOI: 10.2307/48600681. Dopfer, K., Foster, J., & Potts, J. (2004). Micro-meso-macro. Journal of Evolutionary Economics, 14, 263–279. DOI: 10.1007/s00191-004-0193-0. Dopfer, K., & Potts, J. (2008). The general theory of economic evolution. London: Routledge. Dopfer, K., & Potts, J. (2019). Why is evolutionary economics not an empirical science? In F. Gagliardi & D. Gindis (Eds.), Institutions and the evolution of capitalism (pp. 314–326). Cheltenham: Edward Elgar. Edgerton, R. B. (1992). Sick societies: Challenging the myth of primitive harmony. New York: Free Press. Ferguson, K. (2015). Everything is a remix (remastered). YouTube. Retrieved from https://youtu.be/ nJPERZDfyWc Field, A. J. (2008). Why multilevel selection matters. Journal of Bioeconomics, 10, 203–238. DOI: 10. 1007/s10818-007-9018-1. Fomin, I. (2019). Memes, genes, and signs: Semiotics in the conceptual interface of evolutionary biology and memetics. Semiotica, 230, 327–340. DOI: 10.1515/sem-2018-0016. Gagliardi, F., & Gindis, D. (Eds.) (2019). Institutions and the evolution of capitalism: Essays in honour of Geoffrey M. Hodgson. Cheltenham: Edward Elgar. Geels, F. W. (2002). Technological transitions as evolutionary reconfiguration processes: A multi-level perspective and a case-study. Research Policy, 31, 1257–1274. DOI: 10.1016/S0048-7333(02)00062-8. Ginsburg, S., & Jablonka, E. (2019). The evolution of the sensitive soul: Learning and the origins of consciousness. Cambridge, MA: The MIT Press. Godfrey-Smith, P. (2009). Darwinian populations and natural selection. Oxford: Oxford University Press. Henrich, J. (2020). The WEIRDest people in the world. How the West became psychologically peculiar and particularly prosperous. New York: Farrar, Straus and Giroux. Herrmann-Pillath, C. (2010). The economics of identity and creativity: A cultural science approach. New York: Routledge. Herrmann-Pillath, C. (2013). Foundations of economic evolution: A treatise on the natural philosophy of economics. Cheltenham: Edward Elgar. Herrmann-Pillath, C. (2021). Biosemiotic foundations of a Darwinian approach to cultural evolution. Cultural Science Journal, 13, 16–33. DOI: 10.2478/csj-2021-0002. Heylighen, F., & Chielens, K. (2009). Evolution of culture, memetics. In R. A. Meyers (Ed.), Encyclopedia of complexity and systems science (pp. 3205–3220). New York: Springer. Hofhuis, S. T. (2022). Qualitative Darwinism: An evolutionary history of witch-hunting, PhD dissertation, Utrecht University. DOI: https://doi.org/10.33540/1460. Hodgson, G. M. (2011). A philosophical perspective on contemporary evolutionary economics. In J. B. Davis & D. W. Hands (Eds.), The Elgar companion to recent economic methodology (pp. 299–318). Cheltenham: Edward Elgar. Hodgson, G. M. (2015). Conceptualizing capitalism: Institutions, evolution, future. Chicago: The University of Chicago Press. Hodgson, G. M. (2019). Evolutionary economics: Its nature and future. Cambridge: Cambridge University Press. Hodgson, G. M., & Knudsen, T. (2010). Darwin’s conjecture: The search for general principles of social and economic evolution. Chicago: The University of Chicago Press. Hodgson, G. M., & Lamberg, J.-A. (2018). The past and future of evolutionary economics: Some reflections based on new bibliometric evidence. Evolutionary and Institutional Economics Review, 15, 167–187. DOI: 10.1007/s40844-016-0044-3. Johnson, S. (2010). Where good ideas come from: The natural history of innovation. New York: Riverhead Books.

245

Michael P. Schlaile et al. Katz, E. (2006). Rediscovering Gabriel Tarde. Political Communication, 23, 263–270. DOI: 10.1080/ 10584600600808711. Kauffman, S. A. (2000). Investigations. New York: Oxford University Press. Kinnunen, J. (1996). Gabriel Tarde as a founding father of innovation diffusion research. Acta Sociologica, 39, 431–442. DOI: 10.1177/000169939603900404. Kobayashi, D. (2015). Schumpeter as a diffusionist: A new interpretation of Schumpeter’s theory of socio-cultural evolution. Evolutionary and Institutional Economics Review, 12, 265–281. DOI: 10. 1007/s40844-015-0018-x. Koestler, A. (1967). The ghost in the machine. New York: Macmillan. Kronfeldner, M. (2011). Darwinian creativity and memetics. Durham: Acumen. Krul, M. (2018). The new institutionalist economic history of Douglass C. North: A critical interpretation. Cham: Palgrave Macmillan. Law, S. (2011). Believing bullshit: How not to get sucked into an intellectual black hole. New York: Prometheus. Lewens, T. (2020). Cultural evolution. In E. N. Zalta (Ed.), The Stanford encyclopedia of philosophy (Summer 2020 ed.). Retrieved from: https://plato.stanford.edu/archives/sum2020/entries/evolutioncultural/ Markey-Towler, B. (2019). The competition and evolution of ideas in the public sphere: A new foundation for institutional theory. Journal of Institutional Economics, 15, 27–48. DOI: 10.1017/S1 744137418000061. Marsden, P. (2000). Forefathers of memetics: Gabriel Tarde and the laws of imitation. Journal of Memetics - Evolutionary Models of Information Transmission, 4. Retrieved from http://cfpm.org/ jom-emit/2000/vol4/marsden_p.html Mesoudi, A. (2021). Blind and incremental or directed and disruptive? On the nature of novel variation in human cultural evolution. American Philosophical Quarterly, 58, 7–20. DOI: 10.2307/ 48600682. Nelson, R. R., & Winter, S. G. (1982). An evolutionary theory of economic change. Cambridge, MA: The Belknap Press of Harvard University Press. Ostrom, E. (2006). The complexity of rules and how they may evolve over time. In C. Schubert & G. von Wangenheim (Eds.), Evolution and design of institutions (pp. 100–122). London: Routledge. Oxford Dictionaries (undated). Meaning of meme in English. Retrieved from: http://www. oxforddictionaries.com/definition/english/meme. Potts, J. (2019). Innovation Commons: The origin of economic growth. Oxford: Oxford University Press. Rakas, M., & Hain, D. S. (2019). The state of innovation system research: What happens beneath the surface? Research Policy, 48, 103787. DOI: 10.1016/j.respol.2019.04.011. Richerson, P. J., & Christiansen, M. H. (Eds.) (2013). Cultural evolution: Society, technology, language, and religion. Cambridge, MA: The MIT Press. Ridley, M. (2020). How innovation works: And why it flourishes in freedom. New York: HarperCollins. Riedy, C., & Waddock, S. (2022). Imagining transformation: Change agent narratives of sustainable futures. Futures, 142, 103010. DOI: 10.1016/j.futures.2022.103010 Rogers, E. M. (2003). Diffusion of innovations (5th ed.). New York: Simon and Schuster. Rosenberg, A. (2021). The inevitability of a generalized Darwinian theory of behavior, society, and culture. American Philosophical Quarterly, 58, 51–62. DOI: 10.2307/48600685. Roy, D. (2017). Myths about memes. Journal of Bioeconomics, 19, 281–305. DOI: 10.1007/s10818-0179250-2. Runciman, W. G. (2015). Very different, but much the same: The evolution of English society since 1714. Oxford: Oxford University Press. Ruse, M. (1993). Evolution and progress. Trends in Ecology & Evolution, 8, 55–59. DOI: 10.1016/01 69-5347(93)90159-M. Schlaile, M. P., Kask, J., Brewer, J., Bogner, K., Urmetzer, S., & De Witt, A. (2022). Proposing a cultural evolutionary perspective for dedicated innovation systems: Bioeconomy transitions and beyond. Journal of Innovation Economics & Management, 38, 93–118. DOI: 10.3917/jie.pr1.0108. Schlaile, M. P., Mueller, M., Schramm, M., & Pyka, A. (2018). Evolutionary economics, responsible innovation and demand: Making a case for the role of consumers. Philosophy of Management, 17, 7–39. DOI: 10.1007/s40926-017-0054-1.

246

Memes Schlaile, M. P., Urmetzer, S., Blok, V., Andersen, A. D., Timmermans, J., Mueller, M., Fagerberg, J., & Pyka, A. (2017). Innovation systems for transformations towards sustainability? Taking the normative dimension seriously. Sustainability, 9, 2253. DOI: 10.3390/su9122253. Schlaile, M. P. (Ed.) (2021). Memetics and evolutionary economics: To boldly go where no meme has gone before. Cham: Springer. Schmid, H. B. (2004). Evolution by imitation. Gabriel Tarde and the limits of memetics. Distinktion: Scandinavian Journal of Social Theory, 5, 103–118. DOI: 10.1080/1600910X.2004.9672894. Schramm, M. (2008). Ökonomische Moralkulturen. Die Ethik differenter Interessen und der plurale Kapitalismus. Marburg: Metropolis. Schurz, G. (2021). Evolution in nature and culture: Prospects and problems of generalized evolution theory. American Philosophical Quarterly, 58, 95–110. DOI: 10.2307/48600688. Simon, H. A. (1971). Designing organizations for an information-rich world. In M. Greenberger (Ed.), Computers, communication, and the public interest (pp. 37–72). Baltimore, MD: Johns Hopkins Press. Simonton, D. K. (2003). Origins of genius: Darwinian perspectives on creativity. Oxford: Oxford University Press. Sperber, D. (1996). Explaining culture: A naturalistic approach. Oxford: Blackwell. Sperber, D. (2000). An objection to the memetic approach to culture. In: R. Aunger (Ed.), Darwinizing culture: The status of memetics as a science (pp. 163–173). Oxford: Oxford University Press. Stanovich, K. E. (2005). The robot’s rebellion: Finding meaning in the age of Darwin. Chicago: The University of Chicago Press. Tarde, G. (1903). The laws of imitation. New York: Henry Holt. Taymans, A. C. (1950). Tarde and Schumpeter: A similar vision. The Quarterly Journal of Economics, 64, 611–622. DOI: 10.2307/1884391. Urmetzer, S., Schlaile, M. P., Bogner, K. B., Mueller, M., & Pyka, A. (2018). Exploring the dedicated knowledge base of a transformation towards a sustainable bioeconomy. Sustainability, 10, 1694. DOI: 10.3390/su10061694. Velikovsky, J. T. (2016). The holon/parton theory of the unit of culture (or the meme, and narreme). In A. M. Connor, & S. Marks (Eds.), Creative technologies for multidisciplinary applications (pp. 208–246). Hershey, PA: IGI Global. Velikovsky, J. T. (2018). The holon/parton structure of the meme, or the unit of culture. In M. Khosrow-Pour (Ed.), Encyclopedia of information science and technology (4th ed., pp. 4666–4678). Hershey, PA: IGI Global. Veit, W. (2019a). Modeling morality. In Á. Nepomuceno-Fernández, L. Magnani, F. J. SalgueroLamillar, C. Barés-Gómez, & M. Fontane (Eds.), Model-based reasoning in science and technology (pp. 83–102). Cham: Springer. DOI: 10.1007/978-3-030-32722-4_6. Veit, W. (2019b). Model pluralism. Philosophy of the Social Sciences, 50, 91–114. DOI: 10.1177/00483 93119894897. Veit, W. (2023). Model anarchism. THEORIA. DOI: 10.1387/theoria.23849. Veit, W. (2021). Review of Nancy Cartwright’s nature, the artful modeler: lectures on laws, science, how nature arranges the world and how we can arrange it better. Philosophy of Science, 88. DOI: 10.1086/711505. Veit, W. (2022). Health, agency, and the evolution of consciousness. Ph.D. thesis, University of Sydney. https://hdl.handle.net/2123/29836 Veit, W., & Ney, M. (2021). Metaphors in arts and science. European Journal for Philosophy of Science, 11, 44. DOI: 10.1007/s13194-021-00351-y von Bülow, C. (2013). Meme. [English translation of the (German) article “Mem”. In J. Mittelstraß (Ed.), Enzyklopädie Philosophie und Wissenschaftstheorie (2nd ed., Vol. 5, pp. 318–324). Stuttgart: Metzler]. Retrieved via ResearchGate: https://www.researchgate.net/publication/255881715_Meme_encyclopedia_ article von Bülow, C. (2019). Types of selection, selection of types. Unpublished manuscript, retrieved via ResearchGate: 10.13140/RG.2.2.16742.27205. von Hippel, E. (2017). Free innovation. Cambridge, MA: The MIT Press. Wagner, A. (2019). Life finds a way: What evolution teaches us about creativity. New York: Basic Books. Waring, T. M., Kline, M. A., Brooks, J. S., Goff, S. H., Gowdy, J., Janssen, M. A., Smaldino, P. E., & Jacquet, J. (2015). A multilevel evolutionary framework for sustainability analysis. Ecology and Society, 20, 34.

247

Michael P. Schlaile et al. Weber, M. (1930). The Protestant ethic and the spirit of capitalism. (Transl. by T. Parsons; with an introduction by A. Giddens). London: Routledge. Weeks, J., & Galunic, C. (2003). A theory of the cultural evolution of the firm: The intra-organizational ecology of memes. Organization Studies, 24, 1309–1352. DOI: 10.1177/01708406030248005. Williams, G. C. (1966/2019). Adaptation and natural selection. Princeton: Princeton University Press. Williams, G. C. (1992). Natural selection. Oxford University Press. Wilkins J. S., & Bourrat, P. (2022) Replication and reproduction. In E. N. Zalta (Ed.), The Stanford encyclopedia of philosophy (Winter 2022 ed.). Retrieved from: https://plato.stanford.edu/entries/ replication Wilson, D. S., & Gowdy, J. M. (2013). Evolution as a general theoretical framework for economics and public policy. Journal of Economic Behavior & Organization, 90S, S3–S10. DOI: 10.1016/j.jebo.2012. 12.008. Wilson, D. S., & Kirman, A. (Eds.) (2016). Complexity and evolution: Toward a new synthesis for economics. Cambridge, MA: The MIT Press. Wilson, E. O. (1998). Consilience: The unity of knowledge. New York: Random House. Witt, U. (2014). The future of evolutionary economics: Why the modalities of explanation matter. Journal of Institutional Economics, 10, 645–664. DOI: 10.1017/S1744137414000253. Witt, U., & Chai, A. (Eds.) (2019). Understanding economic change: Advances in evolutionary economics. Cambridge: Cambridge University Press.

248

19 THE PATH DEPENDENCE OF KNOWLEDGE AND INNOVATION 1 Cristiano Antonelli and Pier Paolo Patrucco

19.1

Introduction

Path dependence makes an important contribution to the economics tool kit. It has influenced work on the economics of innovation and knowledge generation, fields of investigation that are intrinsically dynamic. This chapter highlights the issues in these disciplines that have been most enriched by the application of path dependence. In economics, the concept of “path dependence” refers to a dynamic property of the allocative stochastic process and can be defined in terms either of the relationship between the process dynamics and the outcome(s) to which it converges or the limiting probability distribution of the stochastic process being considered (David, 1997). The definition of path dependence requires path-dependent and path-independent processes to be distinguished. Path-independent processes include processes whose dynamics guarantee convergence to a unique, globally stable equilibrium configuration or, in the case of stochastic systems, configurations where there exists an invariant (stationary) asymptotic probability distribution that is continuous over the entire space of feasible outcomes. Stochastic systems possessing these latter properties are said to be ergodic and, thus, reversible, and have the ability eventually “to shake free from the influence of their past state(s)” (David, 1997). In contrast, in path-dependent processes the details of the history of the system’s motion do not matter because they cannot affect its asymptotic distribution among states. In fact, path-dependent processes possess a multiplicity of asymptotic distributions, in line, generally, with branching processes, in which the prevailing probabilities of transitions among states are functions of the sequence of the system’s previous transient states. Branching processes subject to local irreversibilities share the property of non-ergodicity, which characterizes these processes’ biological evolution and where speciation constitutes a nonreversible event (David, 1997). Hence, path-dependent processes are considered non-ergodic; thus, they are unable to shake free of their history and must conform to path dependent outcomes (David, 1997). In other words, a path-dependent stochastic process is a process whose asymptotic distribution evolves as a consequence (a function of) the process’s individual history (David, 1997).

DOI: 10.4324/9780429398971-21

249

Cristiano Antonelli and Pier Paolo Patrucco

It follows that path dependence characterizes dynamic systems that are neither completely deterministic (past-dependent) nor purely random (path-independent) in their operation and in which the specific details of history govern their unfolding development (David, 2007). While a past-dependent process is fully deterministic and the initial conditions determine its characteristics and outcomes significantly, a path-dependent process is shaped by the contingent changes occurring in the process whose final outcome is the result of intertemporal externalities and feedbacks within the system integrating the process. (Antonelli, 1997, 2006; Antonelli, Crespi and Scellato, 2012). “History matters” in the case of path dependence since past decisions and contingent events contribute to shaping the economic process and the final outcome, which is not a randomly driven result. At the same time, it is clear that small events occurring at each point in time and space can change the direction and the rate of a path dependent process. Here, the difference between past dependence and path dependence becomes clear. In a past dependent process all the features of the process, its rate and direction, are irreversible and already defined at the onset of the process itself. The trajectory is a past-dependent process (Dosi, 1982). In a pathdependent process, the historic conditions that affect the rate and the direction of the process take place on a continuum which includes both its origins as well as all the steps since its onset. The events occurring at each point in time can affect the future outcomes of the process. In other words, the process is flexible and yet non-ergodic. Actions are influenced by the process memory which keeps adding elements along its way (Antonelli, 1997, 2006). Analysis of the characteristics of non-ergodic processes that are influenced by irreversibility, but are not completely deterministic, allow an exploration of the wide range of possible outcomes of the behaviours of agents that are influenced, but not bound by the initial conditions of the process. At each point in time, agents, aware of the burden of the past, can try to act intentionally and rationally to change the rate and direction of the dynamics. At the same time, small events taking place at the system level also may change the rate and direction of the process. Classification of non-ergodic processes, according to the relevance of the irreversibility, the length of the effective memory and the weight of the effects of small events taking place at each point in time, and their influence on the actions that take place at each point in time, is important and constitutes an area of investigation that deserves exploration. Path dependence introduces the idea that history matters in economics, which challenges the neo-classical approach, which considers both the starting point (initial conditions) and transitory events (accidents) as irrelevant and having no effect on the ultimate outcome of the economic process. According to the neoclassical perspective, economic evolution is subject to random disturbance and is fundamentally unpredictable, and navigation of the various equilibria is ergodic (with no memory), leading to the “best of the achievable worlds” (Antonelli, 2006). Moreover, the theory of path dependence tries to explain “how” history matters when the process of growth and accumulation results in the achievement of one some other equilibrium. Economic processes influenced by history lead to a class of path dependent equilibria which forecloses on the achievement of other classes of equilibria since the costs of switching to another path would result in a substantial reduction of welfare for those that might benefit from the change, making the decision to “switch” irrational. The foreclosure of some equilibria may result in “lock-in” to inferior equilibria, as a consequence, for example, of network externalities (David, 1985). 250

The Path dependence of knowledge and innovation

In the neoclassical approach, which uses the Pareto welfare optimality criterion to judge allocative processes, path dependence is associated to persistent sub-optimal allocations, leading to lock-in phenomena, inertia and market failures. Of course, path dependence can induce inferior equilibria, with not economically or socially optimal solutions persisting and dominating the market, exemplified by the so-called “QWERTY2 economics” (David, 1985), which paved the way to a large stream of work investigating lock-in and stickiness to existing technological paths (Aghion et al., 2015). To sum up, while according to the neoclassical approach and its assumptions, the optimal equilibrium is always achievable and, ultimately, is achieved through an ergodic and random process, according to path dependence theory, the equilibrium achieved may be sub-optimal with respect to potential equilibria reached by alternative paths, but remains the optimal achievable under the historical circumstances and contingencies and nonergodic process that led to it. Although the notion of path dependence has been subject to criticism, especially from neoclassical scholars who struggle with the idea that economic processes could lead to “irremediable errors” (market failures) (Liebowitz and Margolis, 1995), it has exerted a relevant influence on evolutionary economics and has found vast application in the study of the evolution of technologies, institutions, firm strategies, and industry structures (David, 2007). Evolutionary economics identifies path dependence as one of the causes of the persistence of economic desirable activities such as innovation and productivity. In particular, path dependence is a source of true state persistence when past experience, along with contingent factors, has a structural impact on the probability of conducting current and future economic activities. In contrast, a state of spurious persistence arises when the original conditions of a past-dependent process that is underway play an exhaustive causal role in explaining current and future economic activities. Here the notion of meso-economics introduced by Kurt Dopfer helps grasping how the interaction between individual and collective action shapes the path dependence of evolutionary processes: “Significantly, meso builds on the notion of circularity between individual and population. The trajectory dynamic unfolds not as a diffusion of a single valued variable, but rather as a process in which individuals interact with an emergent population in a self-reinforcing way” (Dopfer, 2012: 149). From a policy point of view, while true state persistence has positive connotations and implies that firms’ decisions and creativity as well as policy interventions can have a longterm impact on firms’ activities and outcomes, spurious persistence has negative connotations by implying that any economic activity and their outcome are driven entirely by intrinsic and idiosyncratic features characterizing the actors (firms) performing those activities. The economics of knowledge and innovation has been especially prolific in its attempts to apply the basic tools of the economics of path dependence to obtain a better understanding of the properties of the dynamic processes that take place when the state of the technology and knowledge changes. This is not surprising because, as matter of fact, the notion of path dependence originates from the early analysis of the direction of technological change carried out by Paul David (1975) who showed that the direction of technological change is path dependent because of localized learning processes. Labor abundant countries (UK) did use and learn about (more) labor intensive techniques while capital abundant ones (U.S.) did use and learn about (more) capital intensive techniques. The shape of the frontier of possible innovations is not symmetric but reflects the effects of localized learning and favours the introduction of new (more) labor intensive technologies 251

Cristiano Antonelli and Pier Paolo Patrucco

in labor abundant countries (UK) and (more) capital intensive technologies in capital abundant ones (U.S.). The recent literature on path dependence has been enriched by the attempts to widen the scope of its application by identifying areas of investigation where dynamic processes have a distinctive non-ergodic and yet partly reversible character: i) recent advances in the economics of knowledge and identification of the central role of the limited exhaustibility of knowledge and the non-ergodic character of the knowledge generation processes; ii) revival of the Schumpeterian “creative response”; and iii) analysis of the persistence of innovation. In all of these endeavours, the notion of path dependent processes has been applied successfully with the dual outcome of a tool of investigation that improves our understanding of the economics of knowledge and innovation and enhancement of the original notion of path dependence.

19.2 Path dependence in the generation of knowledge Some recent advances in the economics of knowledge have been related to the properties of knowledge as an economic good. These works pay attention to the limited appropriability and exhaustibility of knowledge and explore the generation of technological knowledge. They identify two characteristics of knowledge which underline the role played by path dependence: i) its reduced exhaustibility; and ii) its intrinsic recombinant nature. Let us reprise them briefly below. Knowledge as an economic good is characterized by both its limited appropriability and its limited exhaustibility. Standard economic goods are affected by obsolescence and typical “wear and use” effects. However, repeated use of knowledge has no effect on and does not reduce its functionality. This limited exhaustibility of knowledge is related to two important properties: extensibility and cumulability. Extensibility refers to the possibility to apply the same knowledge item to repeated and unlimited applications. Extensibility triggers the beneficial effect of the economies of density, according to which average costs decline with the quantity of output using the same input. Cumulability refers to the complementarity among different knowledge vintages. New discoveries are enabled by building on previous discoveries and the implementation of new technological knowledge depends on the specific and highly localized context of application of new scientific knowledge (Arthur, 2007, 2009). In effect, technological knowledge is simultaneously an input into the technology production functions of all the other goods, and in the generation of new knowledge. Its repeated use is sequential: each knowledge item initially is generated and then is used in the technology production function and, in turn, in the generation of new knowledge items. At each point in time, the generation of new technological and scientific knowledge relies on the stock of knowledge generated so far. Existing technological knowledge is an essential and indispensable input to the generation of new technological knowledge. The size and composition of the available existing stock of knowledge at each point in time, has a crucial effect on the amount of new knowledge that can be generated. The larger and the wider the stock of existing knowledge available at each point in time, the easier will be the generation of new technological knowledge (Antonelli, 2018, 2019). The existing stock of knowledge affects the knowledge flows that can be generated at each point in time, in terms of their size and cost and, also, their composition (Patrucco, 2014). The generation of new technological knowledge is based on new combinations of existing knowledge items that increase their scope and functionality. The composition of the existing 252

The Path dependence of knowledge and innovation

stock of knowledge exerts strong non-ergodic effects on the composition of new technological and scientific knowledge flows. Agents with the ability to access different knowledge stocks characterized by knowledge items of varying composition are considered to possess recombinant skills which they apply to different inputs to yield different outputs (Weitzman, 1998). The path dependent properties of the recombinant generation of knowledge characterized by limited appropriability and exhaustibility, seem stronger in the case of technological as opposed to scientific knowledge. Although the generation of scientific knowledge is influenced by the specific composition of the existing stock of knowledge, its outcome tends to be more universal. This applies, in part, to technological knowledge, but technological knowledge, which depends on the specific conditions of its application, is more idiosyncratic and localized in nature. Different pools of scientific knowledge may shape different knowledge generation processes and affect their ability to produce universal items of scientific knowledge. The same scientific knowledge may be produced by differentiated generation processes, which, eventually, converge towards a common outcome that is recognized, validated and shared by all the members of the scientific community. Different pools of knowledge shape not only the process of technological knowledge generation but also the knowledge contents and its specific economic convenience. The path dependent and idiosyncratic effects of the recombinant generation of technological knowledge, in which the existing stock of technological knowledge is an indispensable and specific input, are pervasive because of their strong idiosyncratic nature. The size and composition of the technological knowledge flow that each agent is able to generate at each point in time, are shaped by the size and composition of the stock of knowledge internal to each firm and the size and composition of the stock of knowledge available in the system to which the firm belongs. Here, the limited appropriability of knowledge comes into play. The stock of existing knowledge at each point in time within the confines of each agent spills over and becomes quasi-public with relevant effects on the capability of every other agent to generate new knowledge. The limited appropriability of knowledge means that its limited exhaustibility has relevant effects which affect both the individual agents and the system. Identification of the non-ergodic character of the knowledge generation processes and the pervasive role of the limited exhaustibility of knowledge have led to research into whether dynamic processes display have path dependence characteristics with respect to both their direction and pace.

19.3

Path dependence of the creative response

In his essay “The creative response in economic history”, published in 1947 in the Journal of Economic History, Schumpeter provided a synthesis of his theory of innovation. In a creative response framing, history plays a crucial role on several levels: irreversibility affects decision making at each point in time, yet the agent’s decision making is able to change the rate and direction of his or her actions and performance, generating a dynamic process that is non-ergodic and path dependent. In this framing, firms are credited with limited ability to foresee the future, but need to plan and make decisions with elements of irreversibility. When unexpected changes to product and factor markets expose these firms to out-of-equilibrium conditions, they must try to formulate a response. 253

Cristiano Antonelli and Pier Paolo Patrucco

This response may be adaptive or creative. In the first case, it yields novelty; in the second, it yields productivity enhancing innovation. Firms can respond creatively if they have access to large and wide stocks of both internal and external technological knowledge, which, due to its limited exhaustibility and appropriability, enables production of new knowledge at below equilibrium cost levels and allows productivity enhancing innovations. History matters in a creative response framework on two counts: i) the (weak) irreversibility of the actions taken by the agent at each point in time exerts its effect on its future activities; and ii) the pervasive effects of the size and composition of the stock of knowledge the firm can access at each point in time. The limited exhaustibility of knowledge and its effects in terms of cumulability and extensibility play a central role in the chances that the firm’s response is creative or adaptive. At the same time, it is clear that, at each point in time, firms try to act to allow them cope with unexpected variations in product and factor markets. The introduction of an innovation can change the rate and direction of the process. The firm’s initial conditions affect the outcome, but do not completely shape the effects of its efforts. In turn, the firm’s creative response has non-ergodic effects: i) it triggers unexpected changes in product and factor market conditions, which, in their turn, engender (possible) creative response from other firms; and ii) it changes the size and composition of the stock of knowledge which controls the outcome of the firm’s response, whether adaptive or creative. The dynamics of the creative response is fully path dependent since it combines an appreciation of the effects of irreversibility with the effects of the possible introduction of innovations. History matters for the dynamics of the creative response by affecting the conduct of firms while not producing deterministic processes because it allows account to be taken of the possible actions of all other agents and their consequences at system level. At each point in time, small events change the rate and direction of the process being shaped, but this does not depend exclusively on its initial conditions (Antonelli, 2017). Moreover, the dynamics of the creative response puts the notion of meso-economics into a new light and at the same time benefits from it. The notion of meso-economics challenges the traditional micro-macro dichotomy and stresses the systemic (i.e., meso) dimension of the space into which innovation and knowledge generation do take place as the results of the interactions between individual elements (e.g., entrepreneurs, firms) and macro factors (e.g., knowledge structures and R&D stocks) (Dopfer, 2012). The integration between mesoeconomics and the path dependence of the creative response makes possible to identify the distinction between structural, individual and systemic path dependence. Path dependence is structural when the macro characteristics of the environment into which individual action takes place have a major role in shaping the dynamics. For given levels of reactivity to out-of-equilibrium conditions of product and factor markets- and of the stocks of knowledge internal to each agent, the path dependence of the creative response is mainly determined by the (quasi)public stock of knowledge available in the macro environment as well as by the changes in its structure and in the architecture of interactions and transactions that determine the amount of spillover and their access. Path dependence is instead individual when the characteristics of actors play a stronger role. For given levels and composition of the (quasi)public stock of knowledge available in the system and its access condition, the path-dependent creative response is determined by small changes in the levels of reactivity to out-of-equilibrium conditions of firms and in the size and composition of the stocks of knowledge internal to each agent. 254

The Path dependence of knowledge and innovation

Such a conventional dichotomy between macro- and micro-dimensions of innovation is broken up by the introduction of the systemic (i.e., meso) dimension. Path dependence is finally systemic when it is the result of the interactions between individual (i.e., micro) and structural (i.e., macro) elements. Individual ingenuity and the introduction of new ideas by innovators change the macro conditions of the environment in which individual actors play. Innovation at the micro level does not take place under given conditions but changes these structural conditions themselves, for instance adding to the stock of available knowledge and modifying its structure and composition, as well as the access conditions to it. At the same time, the structure of the stock of available knowledge and the changes in its composition affect future individual creativity and the opportunity to introduce further innovation, modifying both the amount of innovation (rate) and the mix of knowledge underpinning it (direction). The interdependent and recursive connections between micro and macro elements make innovation and the generation of new knowledge a systemic, emergent and open process (Metcalfe, 2002) and reassure the Schumpeterian legacy in the understanding of path dependence.

19.4

Path dependence and the persistence of innovation

Since its inception and early applications, path dependence has been used to explore the direction of non-ergodic processes. The issue of the path dependent character of the rate of dynamic processes was raised later. The concept of persistence has either negative or positive implications, depending on how it is interpreted and the nature of the persistence under consideration. Persistence seems to be related directly to path dependence: history matters not only for shaping the direction of the introduction of technological change but also for the rate of introduction of innovations. There is a reasonable large strand of both theoretical and empirical work on the existence, nature and sources of persistence in innovation activities. From a theoretical point of view, innovation persistence has been defined broadly as the “degree of continuity in innovative activities over time” (Cefis and Orsenigo, 2001), and as including both spurious and true persistence (Manez et al., 2009). Spurious persistence refers to repeated innovations wrought by time-invariant and unobserved firm characteristics, which are largely exogenous to the sequence of innovation, but tend to shape its initial conditions. Therefore, past dependence characterizes spurious persistence and refers to a strong form of non-ergodicity, in which the initial conditions significantly determine process characteristics and outcomes. Although spurious persistence has been less thoroughly explored, it has been linked to a variety of firm-specific factors (Cefis and Orsenigo, 2001), which are distributed heterogeneously across firm, and are stable and hard to change due to their high level of inertia (Stuart and Podolny, 1996). A number of studies identify strategic positioning, corporate culture, research abilities, managerial talent, organizational routines, and organizational and dynamic capabilities as significant firm internal factors affecting the initiation and continuing adoption of innovative practices and activities to ensure persistent innovation (Cefis and Orsenigo, 2001). These factors or characteristics are largely exogenous to the sequence of innovation, but tend to shape its initial conditions. Hence, they induce spurious persistence, which is characterized by past dependence, a strong form of non-ergodicity, where initial conditions 255

Cristiano Antonelli and Pier Paolo Patrucco

significantly determine the characteristics and outcomes of the innovation process (Antonelli, 2006; Antonelli, Crespi, and Scellato, 2012). It should be noted that, since these characteristics often are unobservable, their effect can be difficult to assess empirically. True persistence refers to the manifestation of state dependence, where past innovations have a causal effect on current innovation behaviour, regardless of the continuous influence of unobserved factors. True persistence is an effect of the intertemporal externalities and feedback between subsequent innovation activities and reflects real path dependence. The literature on innovation persistence discusses various mechanisms as the possible causes of true state persistence and proposes four major theoretical explanations for the phenomenon. The first builds on the intuition that “success breeds success”. Scholars argue that, since innovation activities are capital-intensive, risky, and difficult for external financers to assess, firms can face financial constraints which are mitigated by evidence of previous successful innovation (Arrow, 1962; Brown, Fazzari, and Petersen, 2009; Hall, 2002). External funders may see past success as indicating capability and likelihood of more successful innovation activity, which will encourage them to finance the firm. In this case, innovation success breeds innovation success by facilitating access to resources. In addition, previous innovation increases the firm’s profit which can be reinvested in more R&D activity which increases the probability of future successful innovation (Nelson and Winter, 1982; Schumpeter, 1934). The second approach draws on the notion of learning by doing and is rooted in evolutionary theory. It argues that R&D provides increasing returns dynamics and that knowledge is cumulative (Cohen and Levinthal, 1989; Klevorick, Levin, Nelson and Winter, 1995). In this approach, the generation of new knowledge, which drives innovation, involves the firm recombining its prior knowledge with external knowledge to generate new ideas (Rosenberg, 1976; Weitzman, 1998). In this framework, the firm’s accumulated knowledge and experience result in unique and inimitable competences which allow the firm to maintain its innovative performance in line with the current technological trajectory (Dosi and Marengo, 1994; Nelson and Winter, 1982). The third approach adopts an irreversibility perspective based on the idea of the irreversibility of the “sunk costs of R&D investments”. Scholars argue that R&D activities require large start-up costs (e.g., R&D facilities, equipment, hiring and training of scientific and specialized staff), which are largely unrecoverable (Sutton, 1991) and, thus, are a barrier to exit for innovative firms and a barrier to entry for non-innovative firms. This explains the persistence of innovative and non-innovative behaviour among firms (Manez et al., 2009; Sutton, 1991). In this approach, firms that have innovated once are more likely to continue innovation activity because of the low incremental costs related to the internal facilities required to generate new technological knowledge and to innovate (Penrose, 1959). While the above three approaches refer to the main firm internal factors related to true state dependence, the effect of external factors on innovation persistence has also been proposed. These external factors are the quality of local knowledge stocks and so-called Schumpeterian rivalry (Antonelli, Crespi, and Scellato, 2013). Antonelli, Crespi, and Scellato (2013) argue that both inter-industry dynamics, that is, the firm’s access to the stock of local knowledge generated by spillovers from other firms’ innovation activity, and intra-industry dynamics, captured by so-called Schumpeterian rivalry, are significant contributors to innovative activity persistence. Since all the knowledge generated by an industry sector may be useful to all other firms and other activities (Jacobs, 1969), the 256

The Path dependence of knowledge and innovation

higher the level of innovation activity in co-localized firms, the higher the expected level of innovation persistence. At the same time, the higher the level of innovation activity and the more efficient rival firms within the same industry, the higher the likelihood that these individual firms will rely on successful innovation to remain competitive which will strengthen the persistence of innovation (Aghion et al., 2005). External factors shape the context of persistent innovation and exert a path dependent effect on the sequence of innovation. From an empirical perspective, several studies test the existence and nature of innovation persistence, using different innovation proxies including patents, innovation surveys and total factor productivity, and different econometric methods such as dynamic random effects probit models and bias corrected fixed effects models. Patent-based studies find evidence of persistence only in the case of firms with a large number of patents (Cefis, 2003). The main shortcoming of these studies is that they do not allow differentiation between true and spurious state dependence. A more recent stream of work uses data from the Community Innovation Surveys (CIS). For example, Peters (2009) analyses the existence and nature (true vs spurious state dependence) of persistence of innovation activity, employing a dynamic random effects (RE) probit model. About half of those studies that use CIS data and a random effects model find clear evidence of strong persistence, mostly attributed to true state dependence. The remainder find weak evidence or no evidence of the existence of innovation persistence (Antonelli, Crespi, and Scellato, 2012; Roper and Hewitt-Dundas, 2008). Other studies provide differentiated analyses of persistence, related to several types of innovation input (e.g., R&D effort) and output (e.g., product, process and organizational innovations). For example, Hecker and Gantner (2014) highlights that the development of product innovations exhibits significant path dependence, while process and organizational innovations are shaped primarily by time-invariant and unobserved firm characteristics. These authors find evidence, also, of intertemporal complementarities among the various innovation types, such as past process innovation being a determinant of current organizational innovation and vice versa, and past product innovation being a determinant of current process innovation. Some work on innovation persistence uses Total Factor Productivity (TFP) to proxy not for productivity growth, but for innovation. TFP can be a good measure of innovation in contexts of low levels of formalized R&D and patenting activity, where innovation is based mostly on informal research activities, tacit knowledge and learning. In this case, TFP can be the ultimate indicator of the wide array of interrelated effects of the introduction of changes to products, processes, markets, organizations and inputs. The results from studies using TFP to proxy for innovation confirm innovation as a persistent phenomenon within a path dependent process that is shaped by a number of complementarity and contingent factors affecting the local process dynamics (Antonelli, Crespi, and Scellato, 2013, 2015). Finally, the relation between path dependence and innovation persistence has attracted the interest of management of innovation scholars. Path dependence can explain both the persistence of existing institutions, technologies and economic behaviours and the creation of new ones (Vergne and Durand, 2011). In line with this view, an alternative perspective on path dependence has been articulated in which the main elements of path dependence (initial conditions, contingencies and exogenous shocks, self-reinforcing mechanisms and lock-in) are reinterpreted and the idea of path creation is proposed (Garud, Kumaraswamy, and Karnøe, 2010). Subscribers to the notion of path creation state that the initial conditions are not given, but rather are constructed by the actors who mobilize specific sets of events from 257

Cristiano Antonelli and Pier Paolo Patrucco

the past to pursue the current initiative; also, emergent situations are not “contingencies”, but instead, for the embedded actors, facilitate the pursuit of certain courses of action while making others more difficult to pursue; and, finally, in this view, self-reinforcing mechanisms do not just exist, but, instead, are cultivated. Moreover, lock-in is not seen as stickiness to a sub-optimal path or outcome, but as a provisional stabilization within a broader process of structural change, which leaves room for the possibility of creative destruction and creative reaction (Antonelli, 2017), and where those actors with the most to lose proactively tender their creations obsolete in order to survive and introduce further innovations.

19.5

Conclusions

Path-dependent dynamics can yield positive outcomes if it supports the persistence of innovation. However, path dependence dynamics frequently is identified and investigated as a potential source of market failure, as it can induce firms to persist in the allocation of resources to projects that lead to economically and/or socially sub-optimal results. Pathdependent dynamics has been studied, also, as a possible cause of government failure related to the allocation of funds. For example, Antonelli and Crespi (2013) provides an interesting investigation of the causes and the effects of persistence in the discretionary allocation of public subsidies for R&D activities performed by private firms, distinguishing between the so-called vicious and virtuous “Matthew effects”. The virtuous Matthew effect consists of persistence of grant provision to firms that have used previous subsidies to increase their internal knowledge stock, flow of current R&D activities and competence (competence effect). The vicious Matthew effect refers to persistence in the assignment of public subsidies on the basis of reputation and to firms that have reduced their commitment to research following award of a previous subsidy (reputation effect). Based on this distinction, Antonelli and Crespi’s (2013) empirical study demonstrates that persistence is at work in discretionary funding allocation in Italy. However, this funding allocation was not dysfunctional because the competence effect outweighed the reputation effect, which is coherent with the adoption of a “picking-the-winner strategy” by the public authorities. A “picking-the-winner” strategy cannot ensure optimal allocation of public resources and can result in funding of not the best projects, which can lead to lock-in and stickiness to the existing technological path. However, such a strategy could allow public authorities to reduce the costs of government failure associated to selective assignment of public subsidies to R&D activities performed by private firms (Antonelli and Crespi, 2013). Our findings related to the path dependent character of the knowledge generation process and the persistence of innovation have important policy implications. For instance, they suggest that, to be successful, programmes aimed at fostering innovation activities need to consider and match the specific characteristics of the firms and the systems targeted to ensure long run sustained innovation activity.

Notes 1 This chapter contributes to the PRIN 20177J2LS9 research project, which along with Università di Torino and the Collegio Carlo Alberto, provided funding and support. We acknowledge the comments of Kurt Dopfer on preliminary versions of this text.

258

The Path dependence of knowledge and innovation 2 The case of the QWERTY computer keyboard was used by David (1985) to explain the phenomenon of technological lock-in due to network externalities, which made the costs of switching to technological alternatives higher than the costs of continuing to invest in dominant, but inferior technologies.

References Aghion, P., Bloom, N., Blundell, Griffith, R., Howitt, P. (2005). Competition and innovation: An inverted U relationship, Quarterly Journal of Economics, 120, 701–728. Aghion, P., Dechezlepretre, A., Hemous, D., Martin, R., Van Reenen, J. (2015). Carbon taxes, path dependency and directed technical change: Evidence from the auto industry, Journal of Political Economy, 124(1), 1–51. Antonelli, C. (1997). The economics of path dependence in industrial organization, International Journal of Industrial Organization, 15(6), 643–675. Antonelli, C. (2006). Path dependence, localized technological change and the quest for dynamic efficiency, in C. Antonelli, D. Foray, B.H. Hall, and E.W. Steinmuller, (ed.), New Frontiers in the Economics of Innovation and New Technology: Essays in Honour of Paul A. David, Edward Elgar Publishing, Cheltenham, 51–69. Antonelli, C. (2017). Endogenous innovation: The creative response, Economics of Innovation and New Technology, 26(8), 689–718. Antonelli, C. (2018). Knowledge exhaustibility and Schumpeterian growth, Journal of Technology Transfer, 43(3), 779–791. Antonelli, C. (2019). Knowledge as an economic good: Exhaustibility vs appropriability? Journal of Technology Transfer, 44, 647–658. Antonelli, C., Crespi, F. (2013). The “Matthew effect” in R&D public subsidies: The Italian evidence, Technological Forecasting & Social Change, 80, 1523–1534. Antonelli, C., Crespi, F., Scellato, G. (2012). Inside innovation persistence: New evidence from Italian micro-data, Structural Change and Economic Dynamics, 23, 341–353. Antonelli, C., Crespi, F., Scellato, G. (2013). Internal and external factors in innovation persistence, Economics of Innovation and New Technology, 22 (3), 256–280. Antonelli, C., Crespi, F., Scellato, G. (2015), Productivity growth persistence: firm strategies, size and system properties, Small Business Economics (2015) 45, 129–147. Arrow, K.J. (1962). Economic Welfare and The Allocation of Resources for Invention, in Nelson, R.R. (ed.), The Rate and Direction of Inventive Activity, Princeton University Press and NBER, 1962. Arthur, B.W. (2007). The structure of invention. Research Policy, 36(2), 274–287. Arthur, B.W. (2009). The nature of technology: What it is and how it evolves, Free Press, New York. Brown, J.R., Fazzari, S.M., Petersen, B.C. (2009). Financing innovation and growth: Cash flow, external equity, and the 1990s R&D boom, The Journal of Finance, 64 (1), 151–185. Cefis, E. (2003). Is there Persistence in Innovative Activities? International Journal of Industrial Organization, 21(4), 489–515. Cefis, E., Orsenigo, L. (2001). The persistence of innovative activities: A cross-countries and crosssectors comparative analysis, Research Policy, 30(7), 1139–1158. Cohen, W.M., Levinthal, D.A. (1989). Innovation and learning: The two faces of R&D, Economic Journal, 99(397), 569–596. David, P.A. (1975). Technological Choice Innovation and Economic Growth, Cambridge University Press, Cambridge. David, P.A. (1985). Clio and the economics of QWERTY, American Economic Review, 7(2), 332–337. David, P.A. (1997). Path dependence and the quest for historical economics: One more chorus of the ballad of QWERTY, Discussion Papers in Economic and Social History, University of Oxford. David, P.A. (2007). Path dependence: A foundational concept for historical social science, Cliometrica, Journal of Historical Economics and Econometric History, 1(2), 91–114. Dopfer, K. (2012) The origins of meso economics: Schumpeter’s legacy and beyond, Journal of Evolutionary Economics, 22(1), 133–160. Dosi, G. (1982). Technological paradigms and technological trajectories: A suggested interpretation of the determinants and directions of technical change, Research Policy, 11(3), 147–162.

259

Cristiano Antonelli and Pier Paolo Patrucco Dosi, G., Marengo, L. (1994). Towards a Theory of Organizational Competencies, Evolutionary Concepts in Contemporary Economics, Michigan University Press, Ann Arbor, Michigan. Garud, R., Kumaraswamy, A., Karnøe, P. (2010), Path dependence or path creation? Journal of Management Studies, 47(4), 760–777. Hall, B. (2002) The financing of research and development, Oxford Review of Economic Policy, 18(1), 35–51. Hecker, A., Gantner, A. (2014). Path and past dependence of firm innovation, Economics of Innovation and New Technology, 23(5–6), 563–583 Jacobs, J. (1969). The Economy of Cities, Random House, New York. Klevorick, A.K., Levin, R.C., Nelson, R.R., Winter, S.G. (1995). On the source and significance of interindustry differences in technological opportunities, Research Policy, 24(2), 185–205. Liebowitz, S., Margolis, S.E. (1995). Path dependence lock-in and history, Journal of Law Economics and Organization, 11(1), 205–225. Manez, J.A., Rochina-Barrachina, M.E., Sanchis, A., Sanchis, J.A. (2009). The role of sunk costs in the decision to invest in R&D, Journal of Industrial Economics, 57(4), 712–735. Metcalfe, J.S. (2002). Knowledge of growth and the growth of knowledge, Journal of Evolutionary Economics, 12(1–2), 3–15. Nelson, R.R., Winter S.G. (1982). An Evolutionary Theory of Economic Change, Harvard University Press, Cambridge. Patrucco, P.P. (ed.) (2014). The Economics of Knowledge Generation and Distribution: The Role of Interactions in the System Dynamics of Innovation and Growth, Routledge, London. Penrose, E. (1959). The Theory of the Growth of the Firm, Oxford University Press, Oxford. Peters B. (2009). Persistence of innovation: Stylised facts and panel data evidence, Journal of Technology Transfer, 34(2), 226–243 Roper, S., Hewitt-Dundas, N. (2008), Innovation persistence: Survey and case-study evidence, Research Policy, 37(1), 149–162. Rosenberg, N. (1976). Perspectives on Technology, Cambridge University Press, Cambridge. Schumpeter, J.A. (1934). The Theory of Economic Development, Harvard University Press, Cambridge, MA. Schumpeter, J.A. (1947). The creative response in economic history, Journal of Economic History 7, 149–159. Sutton, J. (1991). Sunk Costs and Market Structure: Price Competition, Advertising, and the Evolution of Concentration, MIT Press, Cambridge. Stuart, T.E., Podolny, J.M. (1996). Local search and the evolution of technological capabilities, Strategic Management Journal, 17(S1), 21–38. Vergne, J.P., Durand, R. (2011). The path of most persistence: An evolutionary perspective on path dependence and dynamic capabilities, Organization Studies, 32(3), 365–382 Weitzman, M.L. (1998). Recombinant growth, Quarterly Journal of Economics, 113(2), 331–360.

260

20 EVOLUTIONARY CONSUMER THEORY Andreas Chai and Zakaria Babutsidze

20.1

Introduction

Evolutionary economics is a collective effort to develop a more realistic account of the economy as a complex adaptive system in which knowledge accumulation and the emergence of novelty play a central role (Pasinetti, 1981; Nelson and Winter, 1982; Dopfer et al., 2004). Most of this work has focussed on supply-side learning – technological change undertaken by entrepreneurs, firms, and scientists, but is not the only side that can adapt and learn. Just as the development of new goods requires producers to explore and tinker with the product design and production processes, so consumers come to discover new goods, experiment with their use, and fit them into their everyday lifestyles. They also play an important role in shaping economic evolution as they represent the ultimate selection environment for new goods (Metcalfe, 2001; Witt, 2001; Swann, 2002). A growing number of studies have taken a deeper look at the preference formation process on the demand side. To provide a realistic account of how tastes evolve, scholars in evolutionary consumer theory have considered the cognitive, biological, social, and economic drivers of preference evolution. Evolutionary consumer theory is an effort to understand how tastes evolve according to what consumers learn (Nelson and Consoli, 2010), how they are influenced by social and market environments (Aversi et al., 1999; Babutsidze, 2012; Valente, 2012), as well as their biological origins (Robson, 2001; Saad, 2013). As such, this typically involves moving beyond the standard narrow focus on preference satisfaction. Instead, a broader focus on preference formation is developed where explicit consideration is given to how preferences are influenced by rising living standards associated with economic development. Moreover, evolutionary consumer theory has also provided insights into the role that consumers play in the innovation process by stimulating research and in some cases directly contributing to emergence of innovations. On the meso level, evolutionary consumer theory has also shed light on how the evolving character of demand can stimulate structural change in the industrial compositions of economies. Taken together, these contributions deliver insights into how knowledge accumulation on the supply side can shape and also be influenced by patterns of consumer knowledge accumulation.

DOI: 10.4324/9780429398971-22

261

Andreas Chai and Zakaria Babutsidze

20.2 Modelling bounded rationality The first contribution of evolutionary consumer theory lies in several models that provide a more realistic account of consumer behaviour. Economic analysis is fundamentally the analysis of human behaviour. However, a newcomer to economics who peruses its work may be a little confused. In contrast to the natural sciences, economists seem to prefer minimalist and abstract models of behaviour and show little interest in the voluminous amounts of work undertaken on behaviour, development and influences available in other disciplines such as psychology, sociology, anthropology, and biology. The consumer in standard economic theory is simply not human (Veblen, 1898; Metcalfe, 2001). Rather, it is some Olympian ‘lightning calculator’ who, despite his wisdom, is not seen as an agent of real change but instead as an impassive arbitrator of markets (Swann, 2002). Typically, the spending behaviour of an entire population is modelled using a representative agent approach in which income and price are the sole explanatory variables (Deaton and Muellbauer, 1980). These foundations stem from the notion of perfect rationality. This concept presumes that the best way to depict economic system’s behaviour is to assume that agents are fully rational and have perfect foresight. This assumption delivers a simply and tractable model of human behaviour that is usually justified by the argument that economic agents who are not perfectly rational (i.e. behave sub-optimally) will be driven out of the system by competitive forces in the long run. Therefore, in equilibrium, we can model the system by studying the collection of perfectly rational agents. Given that what we study is a model, we need not replicate the behaviour of each individual agent, but rather behaviour of the system which would be equivalent to that populated by perfectly rational agents. Perfect rationality was developed to study supply side production decisions under free market theory and it relied on competitive forces to drive poor quality (or irrational) producers out of the market. Over time, the analytic convenience of the approach was quickly mirrored into the consumer theory. However, as pointed out by Valente (2012), the “as if” argument applicable to competitive firms does not apply to consumers – we cannot expect non-rational consumer behaviour to be driven out of the market by competitive forces. Indeed, it could easily be conjectured that markets could encourage the propagation of non-rational behaviour for the purposes of profit maximisation (Akerlof and Shiller, 2015). There are no good reasons to conjecture that a model inhabited by perfectly rational consumers will reasonably approximate the society inhabited by (many) non-rational consumers. This suspicion is amplified by empirical regularities about human cognitive capabilities (Campbell and Kirmani, 2000), memory limitations (Braun, 1999) and slow learning processes (Israel, 2005). Hence, the existing obsession with perfect rationality, prevalent in mainstream economics (on the demand side), is likely to strongly inhibit our capacity to study complex social systems. As a result, there is a greater case for developing a realistic view of behaviour when considering consumers rather than suppliers. Many evolutionary economists have been actively departing from the assumption of perfect rationality on the consumer side. This alleviates requirements of human cognitive capacity and perfect foresight in highly uncertain circumstances. It allows for more realistic micro-foundations that would be related to empirical evidence on consumer behaviour (Babutsidze, 2012). It also allows us to study the heterogeneity with more rigor as non-rationality does not require identical consumers.

262

Evolutionary consumer theory

However, one important challenge here is that one can depart from perfect rationality in many ways. There is only one way of being rational, but infinite ways of being irrational. One general term that has been applied to these numerous departures has been the concept of bounded rationality, which highlights limits of the economic agents’ mental capabilities (Simon, 1956; Earl, 1986).1 However, this concept also subsumes departures in terms of varying objective functions and in terms of the methods for satisfying these objectives. For example, orthodox economics takes consumer preferences as given (and constant) and leaves the question of preference formation to other social sciences. This would indeed be productive if preferences did not change and co-evolve with other variables in the system (Saviotti, 2001). However, as we cannot have tastes defined for any potential product that might emerge in the future, we need to incorporate consumer taste formation and change in evolutionary theorizing. Although elements of evolutionary theorizing on consumer behaviour have been present in literature as early as the seminal contribution by Nelson and Winter (1982) (most notably Pasinetti, 1981; Cowan et al., 1997; Aversi et al., 1999), major explicit attempts to form a comprehensive evolutionary model of consumer behaviour are tied to a special issue in the Journal of Evolutionary Economics, edited by Ulrich Witt in 2001. Scholars contributing to this special issue were dissatisfied by staggering inertness of consumers in standard economic growth models. Contributions strived for a more reasonable treatment of knowledge in growth models (Langlois, 2001), as well as explicit incorporation of time in consumption activities (Metcalfe, 2001).2 Perhaps the two most important contributions emerging from special issue were ones by Witt (2001) and Saviotti (2001). Witt (2001) proposed to start theorizing from antecedents to classical preferences – wants.3 The main aim of this seminal contribution was to understand the long run dynamics through which the economic system in developed countries has been escaping the satiation in consumption (discussed in the next section). It was theorized that systematic increase in consumption was due to development of new wants through cognitive and non-cognitive learning processes. This put the (co-)construction of consumer preferences at the top of the evolutionary modelling agenda. Complementary to this approach, Saviotti (2001) has proposed to look at (dramatic) increase in the variety of supplied goods as a qualitative transformation of the economy that required significant adjustments from consumers. Taking these two contributions together, evolutionary economists had two important ingredients of evolutionary consumption theory – product innovation and (slow) learning. These building blocks were further complemented by Nelson and Consoli (2010), who convincingly argued that utility maximization was not required to obtain the downwardsloping demand curve, which is an ultimate empirical regularity to be replicated by a sensible model of consumer behaviour. Nelson and Consoli (2010) accentuate the importance of time it takes for consumers to react to changing market circumstances. They highlight the importance of changes in absolute prices and argue that the speed of substitution between products following price changes are different depending on which price has changed. If the price of the product consumed by the household has changed, the transition to the new consumption bundle is faster compared to the case when it is the price of the substitute product that has changed. This highlights the differential timescales on which consumer learning processes might operate. Nelson and Consoli (2010) propose a theoretical framework where consumer decisions are implemented as routines which are only marginally altered with time. As a consequence, the proposed framework exhibits heterogeneity, inertia 263

Andreas Chai and Zakaria Babutsidze

and path-dependence in consumption patterns which are also important building blocks of evolutionary consumption theory. As seldom-revised consumption routines and rules of thumb are used as guiding principles for consumer decisions in evolutionary theorizing, most of the evolutionary models are characterized by properties of path dependence and systemic inertia (FatasVillafranca et al., 2009). This inertia could result into lock-in effect which could very well be sub-optimal like the technological lock-in by Arthur (1989). Habituation constitutes a dominant pattern of behavior (Babutsidze and Cowan, 2014) that needs to be punctuated by innovative activities corresponding to purchase of new products and services. Such deviations are usually modeled as random errors in (currently) optimal (and thus stable) consumer behavior. However, thanks to these “errors” consumers can continue the exploration process and discover new (potentially superior) activities for which they can develop tastes. This process could quick compound across the population of consumers and adoption trajectories for new goods (Dolfsma and Leydesdorff, 2009). The intensity of exploration of surroundings is usually modeled as the function of current level of satisfaction. Thus, adoption to novelty is more rapid in market segments that leave consumers dissatisfied. Looking at consumer behaviour through the lens of sticky, heterogeneous routines over the changing product set available on the market begs the following question: under which circumstances can we expect consumers to take cold-headed, (more, but still less than fully) rationally calculated actions and when do we expect them to follow routines that are occasionally revised? Empirical evidence (as well as psychological theorizing) suggests that following consumption routines are particularly prevalent when consumers are negotiating frequently repeated purchases, and/or spending of relatively small share of their disposable income (Babutsidze, 2012). When purchase decisions concern larger stakes (i.e. purchasing a house or a car) and unfamiliar products more calculations take place (Shugan, 2006). However, even in the case of certain unfamiliar products, due to imperfect information, consumers might be forced to resort to imperfect behavioural routines – for example in case of experience goods like movies (Babutsidze and Valente, 2019). Another conceptual approach for evolutionary consumption theory is to focus on choice as a mental phenomenon and consider the decision criteria through which consumers come to make their choices (Earl, 1986; Valente, 2012). Earl (1986) emphasizes how economic alternatives are the products of individual acts of creative imagination. To navigate choice in an uncertain world, consumers develop consumption strategies that build complementarities between different aspects of their consumption routines – which label ‘the consumer lifestyle’. Valente (2012) defines consumer preferences as an ordered set of product characteristics. Such formulation is general enough to guide consumer decisions over a large set of products that have not yet appeared on the market. Under this formulation the modeler can forecast consumer reaction to a potential new product with a certain set of characteristics. This simplifies the study of innovative activities coming from supply side. However, a more recent contribution by El Qaoumi et al. (2017) shows that consumer learning in the context of novelty can be reformulated as the expansion of the set of product characteristics. Taking into account this possibility, Valente’s model could be extended by modeling the change in the choice criteria as consumers discover new products and learn about new dimensions to evaluate. Naturally, this extension would involve consumption inertia as learning is not instantaneous. 264

Evolutionary consumer theory

20.3

Accounting for consumption laws

A second important line of research in evolutionary consumer theory is to account for empirical regularities observed in consumption patterns (Pasinetti, 1981; Aversi et al., 1999; Chai and Moneta, 2012; Lades, 2013; Witt, 2017). There is a set of well-known empirical regularities describing the way household spending patterns tend to evolve as household income grows (Kindleberger, 1989). A classic example is Engel’s law (Lades, 2013). To account for these long run trends, scholars have noted that one cannot just consider the evolution of the consumer’s decision-making routines, but more deeply examine the biological and social determinants of demand. This has been tackled by theorizing about the underlying wants that motivate consumption and considering what changes have taken place in the way these wants are satisfied (Witt, 2001; Lades, 2013). Wants are defined as states of being that motivate consumers to engage in behavior. Much evidence suggests that biological evolution has left humans predisposed to engaging in certain behaviors that deliver primary reinforcement such as food, sex, peer recognition and health. By studying the evolving connection between innate wants and the set of goods that can satisfy those wants, it becomes possible to arrive at a more fundamental account of the non-homothetic nature of consumer preferences. The most prominent example of this wants-based analysis of consumption patterns is the case of food consumption used to satisfy the need for hunger (Ruprecht, 2005; Manig and Moneta, 2014). This underlying want for food is homeostatic in nature and suggests that a satiation level of food exists. Manig and Moneta (2014) provide evidence for satiation in food consumption. In their cross-sectional empirical investigation of contemporary Russian food spending patterns, the authors examine the relationship between calorie consumption and income (see inter alia Bouis and Haddad, 1992). Beyond food, satiation dynamics are observed in other consumption domains, including alcohol (Volland, 2012), washing machines (Woersdorfer, 2010) and shoes (Baudisch, 2007). Other studies have sought empirical evidence for the satiation hypothesis by investigating the shape of Engel curves using data on household expenditure (Kaus, 2013, Bruns and Moneta, 2017). An open question is how social environments may influence satiation levels (Woersdorfer, 2010). Cordes (2009) and Lades (2013) suggest that for certain socially orientated needs such as social esteem, changes in the peer groups and the affluence level of peer groups could generate a rise in the satiation level of expenditure required to satisfy these needs. This effort is close to other studies that have considered status seeking motives and its impact on household spending (Frank, 1985; Kaus, 2013; Heffetz, 2011; Hopkins and Kornienko, 2004). A key insight from this literature is the observation that changes in the income distribution could trigger important changes in the intensity of conspicuous spending patterns among consumers. If consumers do assess their welfare by comparing their consumption level to their neighbors’ consumption levels, scholar need to confront the possibility that welfare levels will not rise always and may in fact diminish as consumption levels rise (Clark et al., 2008; Easterlin et al., 2010).

20.4 Insights from the population level A key aspect of evolutionary economics is population thinking that recognizes both heterogeneity on the individual level and group level properties can have important implications for the rate of economic evolution (Metcalfe, 2001). By recognizing heterogeneity in

265

Andreas Chai and Zakaria Babutsidze

consumption behavior and demand patterns of consumers, evolutionary consumer theory has developed insights into emergent properties that can influence the rise of innovations. A population perspective was recognized to be important based on insight about the uneven nature in which new products tends to be adopted across a population of consumers (Bass, 1969; Janssen and Jager, 2002; Kiesling et al., 2012). Neoclassical models tend to assume frictionless markets that enable instantaneous adjustments to changing market circumstances. In reality, adjustments are more likely to be slow as it requires demand-side adaptation, and through which consumes come to learn about new goods and how to use them. Thus preferences, decisions and demand are evolving alongside changes in available product set and their attributes (i.e. price and quality). This naturally creates diffusion patterns for new products which constitutes an important aspect of evolutionary theorizing. The impetus for such models originated in early mathematical treatment of the diffusion process was developed by Bass (1969). This model acknowledged the importance of two distinct roles among consumer population – innovators and imitators – and modeled the new product diffusion as a differential equation under the assumption of continuous random interaction among consumers. A similar idea of unstructured interaction was adopted by early evolutionary consumption models, most notably Aversi et al. (1999) who present a computation model of consumer behavior based on genetic algorithm. This is a more micro-founded consumer behavior model where consumer heterogeneity is treated more carefully. Yet, like Bass (1969) this model made no attempt to model social interactions on the demand side. An early contribution to the literature explicitly modeling interaction structure is due to Cowan et al. (1997). The authors model social interaction among strictly defined groups of consumers based on their willingness to converge to a distinct consumption pattern. Cowan et al. (1997) succeed in tying certain stable consumption patterns to the properties of social interactions. Most importantly, the authors are able to generate heterogeneity in consumption patterns in ex-ante homogeneous consumption groups. A later contribution by Babutsidze and Cowan (2014) model individual (rather than group) interactions and shows that stable consumption patterns can emerge without exogenously defined community boundaries. In this setup, consumption communities emerge endogenously. Another class of models deals with interaction not at the system or a group level, but on an individual level (Babutsidze and Cowan (2014) discussed in the previous paragraph belongs to this class of models). These models impose certain structure on individual interactions using explicit networks. The simplest (and the earliest) of these studies model new product diffusion as a percolation process in a homogeneous (usually two-dimensional) lattice (Hohnisch et al., 2008; Cantono and Silverberg, 2009). An important advantage of these models is the fact that they can tie emergent properties of the model (i.e. realized aggregate features of the market which do not have explicit precursors in modelled micro behavior) to the nature of interaction among consumers that constitutes the base for learning in the evolutionary framework. Later contributions examine social network structures that are closer to networks observed in the field. These contributions usually tie economic outcomes to the structure of consumer interactions rather than to the nature of these interactions. In this framework, speeds of learning and adjustment, and therefore diffusion patterns, are moderated by topologies of network over which consumers interact. These models are most suited for modeling social learning and information exchange through word-of-mouth interactions. Using this framework, Zeppini and Frenken (2018) investigate economic efficiency of 266

Evolutionary consumer theory

diffusion processes resulted from a number of different network structures and Babutsidze (2018) investigates implications for producers who are striving for further and faster diffusion of their products. Modeling consumer interaction on population level is useful for studying not only innovation diffusion patterns within (parts of) population, but also conditions under which innovations emerge.

20.5

Demand and innovation

Apart from acting as a selection environment that determines success of failure of novelty, there has also been effort to understand how consumer expertise and acts of creativity by consumers can stimulate innovation (Bianchi, 1998; Mueller et al., 2015; Schlaile et al., 2018). Such work can help generate a more comprehensive understanding of how novelty emerges in the context of an evolving economic system (Dopfer et al., 2004; Witt, 2008; Harper and Endres, 2012). This can be divided into two broad parts: indirect approach and direct approach. The first set of studies (indirect approach) consider how the extent of the market incentivizes suppliers to engage in more innovative activity. Schmookler’s demand-pull hypothesis states that inventive activity within industries is responsive to the pull of demand. Demand growth increases the expected future profits from innovation (Schmookler, 1960; Scherer, 1982). Empirical evidence for this effect is relatively mixed (Kleinkrecht and Verspagen, 1990; Fontana and Guerzoni, 2008). This is likely attributable to the complex nature of the innovation process that requires adequate levels of R&D capability at the micro level as well as the correct institutional regimes and incentive structures at the macro level (Nelson and Rosenberg, 1993; Fagerberg and Srholec, 2008). A second line of research examines how the heterogeneity of demand may help influence innovative activity. Expert consumers can help firms develop new products. It is argued that markets with a higher number of specialized consumers have a higher probability of witnessing the introduction of novelties which have been co-developed with consumers (Jeppsen and Molin, 2003; von Hippel, 2005). Segments of specialized consumer preferences can also create ‘niche markets’ which can potentially sustain new prototype goods that may not be competitive in the wider mass market (Windrum, 2005; Nieman and Vavra, 2019). Thereby, the existence of niche markets can play a critical role in industry evolution (Saviotti, 1996; Guerzoni, 2010; Malerba et al., 2007). As such, the heterogeneous character of demand is considered a type of potential resource or ‘capability’ that firms can utilize in the innovation process. On the industry level, the diverse nature of consumer preferences and the associated presence of niche markets can play a critical role in influencing how market competition fosters the emergence of dominant designs within an industry (Saviotti, 1996; Bresnahan and Gambardella, 1998; Lipsey et al., 2005; Malerba et al., 2007). Taken together, these contributions have highlighted when considering how consumer demand may foster the emergence of innovations, how it is not only important to consider the depth of demand, but also the heterogeneity of demand across different segments and the unique preferences within each of these segments that can play critical role in the innovation process. Beyond innovative activity responding to changes in market depth and the growth rate of demand, another group of scholars (direct approach) examine how innovative activity is shaped by what entrepreneurs perceived to be the needs of consumers (Mowrey and Rosenberg, 1979; von Hippel, 2005). A variety of studies have examined the role that expert consumers play in co-developing novel products and services (Jeppsen and Molin, 2003; 267

Andreas Chai and Zakaria Babutsidze

von Hippel, 2005; van den Ende and Dolfsma, 2005). This ‘consumers as producers’ perspective makes the valid point that much of the time inventors and entrepreneurs responsible for introducing novelty are also (highly specialized) users, and their interest in developing innovation may be personal as well as pecuniary. Markets in which there are a higher number of specialized consumers have a higher probability of witnessing the introduction of novelties which have been co-developed with consumers. Of course, whether or not co-development takes place also depends on whether consumers and producers have the proper means to cooperate with each other, whether the goods and services in question can be developed on a small scale, or whether the economic system is generally open to such entrepreneurial activity (van den Ende and Dolfsma, 2005).

20.6

Consumer demand and structural change

How does the evolving character of demand impact economic growth? The industrial composition of the economy tends to undergo important structural changes as it grows. Scholars have long conjectured that this phenomenon may be linked to systematic changes in the character of household spending (Engel, 1857; Kuznets, 1973). Many scholars posit that the non-homothetic nature of consumer demand plays an important role in driving these changes (Pasinetti, 1981; Bertola et al., 2006; Ciarli et al., 2010; Saviotti and Pyka, 2013). The chief mechanism through which non-homothetic preferences impacts the growth rate of industries is the growth rate of market demand which tends to slow after reaching a set threshold due to satiation of demand for certain items (Aoki 2002,Aversi 1999; Metcalfe et al., 2006; Desmarchelier et al., 2017). A second theory is the ‘escaping satiation’ hypothesis which states that as household incomes rise and a greater proportion of consumers reaches the saturation level of spending on a good, the associated slowdown in demand growth stimulates inventive activity (Witt, 2001; Fatas-Villafranca and Saura-Bacaicoa, 2004). This suggests that the characteristics of goods will evolve and adopt new features considering demand saturation. If successful, such product innovations effectively push the saturation level of overall spending to a higher level. As a result, the shape of Engel Curves could exhibit systematic instability over time as slowdowns in demand trigger product innovations (Moneta and Chai 2013; Moneta 2013). Some preliminary empirical evidence consistent with the notion of satiation-escape can be found by studying the co-movement of the Engel Curves’ satiation level and the average income level of the household population (Kaus, 2013; Moneta and Chai 2013). A handful of other studies have further explored how other aspects of consumer preferences and consumption patterns may have macro-economic impacts. For example, Ciarli et al. (2010) and Lorentz et al. (2016) also consider how the selectivity of consumer preferences (choosing goods that fulfill a certain performance threshold) have a macro impact. These authors also explicitly consider classes of consumers which recognize the heterogeneous nature of consumer demand. A number of other studies have also explicitly explored the emergent nature of these classes and how they may evolve as result of paradigm shifts (Fatás‐Villafranca et al., 2007; Rengs & Scholz-Wäckerle, 2019).

20.7 The coevolution of demand and supply One of the most fundamental and open questions in evolutionary consumer theory is to understand the precise extent consumer preferences are endogenously influenced by economic 268

Evolutionary consumer theory

growth and development (Bowles, 1998). If indeed economic conditions shape preferences, this challenges the notion of consumer sovereignty and the principle that all economic activity should be directed to the satisfaction consumer preferences (Lintott, 1998; Røpke, 1999). Answering the question of how consumer preferences co-evolve with technological progress could help unlock a new market-based theory of endogenous economic growth that is focused on the interaction between consumers and suppliers. Dopfer (2016) proposes a new approach to analyze the interaction between demand and supply by studying the rules of consumption and the rules of production. As such, the market can be understood as an institution in which agents carrying demand rules and supply rules interact with each other. From the Schumpeterian perspective, a proper microeconomic foundation for understanding how a novelty emerges in economic systems involves understanding why and under what conditions individual agents are more or less inclined to search for and discover novelty (Witt, 2008; Foster and Metcalfe, 2012; Harper and Endres 2012). ‘Agents’ here refers to not just entrepreneurs, but also to consumers with their inclination to search, discover and adopt new goods. To construct a more complete picture of self-transformation, more consideration must be given to how consumer demand co-evolves with new technologies and changing market conditions (Fatas-Villafranca et al., 2019). In doing so, the focus of analysis shifts back on the market interaction between consumers and suppliers. According to Schumpeter, this interaction is not just a mechanism for coordinating economic activity but the core engine of economic growth (Schumpeter, 1934). For example, the satiation process (discussed above) represents a self-transformation of the system. A slowdown in consumption expenditure due to the satiation of the want for food, which was itself caused by improving advances in food production technologies, has led to innovation in the food industries that look to overcome this slowdown by modifying goods to appeal to new wants. As a result, evolutionary theorists have recently made a headway combining the two processes and modeling co-evolution of demand and supply when discussing innovation. Early contribution by Ida (2010) discusses the co-dependence between preference and product quality evolution using the model inspired by research in population ecology. The author derives conditions under which the co-evolutionary process becomes self-reinforcing and locks into an ever-increasing (runaway) dynamics. Using an agent-based framework, Safarzynska and Van den Bergh (2010) model interdependence between consumer preferences and its impact on firm’s innovation processes. This results in a co-evolutionary process that is suitable to study policy effects on the ecological transition. Saviotti and Pyka (2013) tackle the challenging task of evaluating the importance of demand-side dynamics on innovation processes and economic development. They demonstrate that a model that is devoid of dynamic responses on the demand side is incapable of replicating important features of economic development. More recent contribution to this stream of research goes further into technicalities. Mueller et al. (2015) present a model that results into endogenous dynamic market segmentation and emergence of stable niches. In a related framework, Babutsidze (2017) shows how market characteristics influence the link between consumer behavior and producer’s innovative activities.

20.8

Conclusion

Despite the diversity of contributions across the micro, meso and macro domains, Evolutionary consumer economics can be viewed as a contribution to the general goal of 269

Andreas Chai and Zakaria Babutsidze

developing a realistic understanding of how consumers adapt to changes in constantly evolving systems. This body of work has contributed to developing a more realistic understanding of i) consumer decision making, ii) the process of preference formation, iii) the role of consumers in the innovation process, iv) the diffusion of innovations among heterogenous consumers, and v) the co-evolution of demand and supply. Taken together, these insights can help policymakers understand the path-dependent nature of consumption and design policies aimed at breaking current consumption habits and transitioning toward a more sustainable future.

Notes 1 An alternative approach is to consider consumers taking decisions based on what they consider is optimal. This approach maintains rationality of individual consumers but acknowledges the possibility of imperfect knowledge ( Markey-Towler, 2018). 2 Due to space constraints, we do not cover a number of contributions that have worked on developing a more realistic approach to consider the role that time play in the optimal sequencing of consumption activities. As noted by Gossen (1854) and Georgescu-Roegen (1983), this has fundamental implications for understanding consumption dynamics. For further work see Steedman (2001), Nisticò (2005), Bianchi (2008) and Fellner and Seidl (2015). 3 Several efforts to develop a wants- or needs-based approach to consumption exist. Among others, these include Menger (1871), Lavoie (1994), Jackson et al. (2004), Saad (2007), Griskevicius et al. (2013).

References Akerlof, G. A., & Shiller, R. J. (2015) Phishing for Phools: The Economics of Manipulation and Deception, Princeton University Press. Aoki, M., & Yoshikawa, H. (2002) Demand saturation-creation and economic growth. Journal of Economic Behavior and Organization 48:127–154. Arthur, W. B. (1989) Competing technologies, increasing returns, and lock-in by historical events. The Economic Journal 99:116–13. Aversi, R., Dosi, G., Fagoilo, G., Meacci, M., & Olivettu, C. (1999) Demand dynamics with socially evolving preferences. Industrial and Corporate Change 8(2): 353–408. Bass, F. (1969) A new product growth for model consumer durables. Management Science 15:215–227. Babutsidze, Z. (2012) Hoe do consumers make choices? A survey of evidence. Journal of Economic Surveys 26:752–762. Babutsidze Z. (2017) Consumer interaction and innovation incentives. International Journal of Computational Economics and Econometrics 7(3):280–301. Babutsidze Z. (2018) The rise of electronic social networks and implications for advertisers. Technological Forecasting and Social Change 137:27–39 Babutsidze, Z., & Cowan, R. (2014) Showing of telling? Local interaction and organization of behavior. Journal of Economic Interaction and Coordination 9:151–181. Babutsidze, Z., & Valente, M. (2019) Trick of the tail: The role of social networks in shaping distributional properties of experience-good markets. Industrial and Corporate Change 28(3):459–475 Bertola, G., Foellmi, R., & Zweimüller, J. (2006) Income Distribution in Macroeconomic Models, Cambridge, MA, Princeton University Press. Bianchi, M. (1998). The taste for novelty and novel tastes. The active consumer: Novelty and surprise in consumer choice 20:64. Bianchi, M. (2008) Time and preferences in cultural consumption. Value and valuation in art and culture, in M. Hutter, & D. Throsby (eds.) Value and Valuation in Art and Culture, Cambridge, Cambridge University Press. pp.236–260. Bowles, S. (1998) Endogenous preferences: The cultural consequences of markets and other economic institutions. Journal of economic literature 36.1:75–111.

270

Evolutionary consumer theory Bouis, H. & Haddad, L. (1992) Are estimates of calorie-income elasticities too high?: A recalibration of the plausible range. Journal of Development Economics 39:333–364. Braun, K. (1999) Postexperience advertising effects on consumer memory. Journal of Consumer Research 25:319–334. Bresnahan, T. Y., & Gambardella, A. (1998) The Division of Inventive Labor and the Extent of the Market, in Helpman (Ed.) General Purpose Technologies and Economic Growth. Bruns, S. B., & Moneta, A. (2017). Intertemporal propensity to consume. Journal of Evolutionary Economics 27:295–314. Campbell, M., & Kirmani, A. (2000) Consumers’ use of persuasion knowledge: The effects of accessibility and cognitive capacity on perceptions of an influence agent. Journal of Consumer Research 27:69–83. Cantono, S., & Silverberg, G. (2009) A percolation model of eco-innovation diffusion: The relationship between diffusion, learning economies and subsidies. Technological Forecasting and Social Change 76:487–496. Chai, A., & Moneta, A. (2012) Back to Engel? Some evidence for the hierarchy of needs. Journal of Evolutionary Economics 22:649–676. Ciarli, T., Lorentz, A., Savona, M., & Valente, M. (2010) The effect of consumption and production structure on growth and distribution. A micro to macro model. Metroeconomica 61(1):180–218. Clark, A. E., Frijters, P., & Shields, M. A. (2008) Relative income, happiness, and utility: An explanation for the Easterlin paradox and other puzzles. Journal of Economic literature 46(1):95–144. Cordes, C. (2009) Changing your role models: Social learning and the Engel curve. The Journal of Socio-Economics 38(6):957–965. Cowan, R., Cowan, W., & Swann, P. (1997) A model of demand with interaction among consumers. International Journal of Industrial Organization 15:711–732. Deaton, A., & Muellbauer, J. (1980) An almost ideal demand system. American Economic Review 70:312–326. Desmarchelier, B., Djellal, F., & Gallouj, F. (2017). Economic growth, business cycles and products variety: Exploring the role of demand satiety. Journal of Evolutionary Economics 27(3):503–529. Dolfsma, W., & Leydesdorff, L. (2009). Lock-in and break-out from technological trajectories: Modelling and policy implications. Technological forecasting and social change 76(7):932–941. Dopfer, K., Foster, J., & Potts, J. (2004) Micro-meso-macro. Journal of evolutionary economics 14(3):263–279. Dopfer, K. (2016) Evolutionary Economics, in: G. Faccarello, & H. D. Kurz (eds.) Handbook of the History of Economic Analysis, Vol. II, Cheltenham, Edward Elgar, 2016, pp. 175–193. Earl, P. (1986) Lifestyle Economics: Consumer Behaviour in a Turbulent World, Sussex, Wheatsheaf Books. Easterlin, R. A., McVey, L. A., Switek, M., Sawangfa, O., & Zweig, J. S. (2010) The happiness–income paradox revisited. Proceedings of the National Academy of Sciences 107(52):22463–22468. El Qaoumi, K., Le Masson, P., Weil, B., & Ün, A. (2017). Testing Evolutionary Theory of Household Consumption Behavior in the case of Novelty – Product characteristics approach. Journal of Evolutionary Economics 28:437–460. Engel, E. (1857) Die lebenskosten Belgischer arbeiter-familien fruher und jetzt. International Statistical Institute Bulletin 9:1–74. Fagerberg, J., & Srholec, M. (2008) National innovation systems, capabilities and economic development. Research Policy 9:1417–1435. Fatas-Villafranca, F., & Saura-Bacaicoa, D. (2004) Understanding the demand-side of economic change: A contribution to formal evolutionary theorizing. Economics of Innovation and New Technology 13(8):695–716. Fatás‐Villafranca, F., Saura, D., & Vázquez, F. J. (2007) Emulation, prevention and social interaction in consumption dynamics. Metroeconomica 58(4):582–608. Fatás-Villafranca, F., Saura, D., & Vazquez, F. J. (2009) Diversity, persistence and chaos in consumption patterns. Journal of bioeconomics 11(1):43–63. Fatas-Villafranca, F., Fernández-Márquez, C. M., & Vázquez, F. J. (2019) Consumer social learning and industrial dynamics. Economics of Innovation and New Technology 28(2):119–141. Fellner, W., & Seidl, R. (2015) Satiated consumers: Allocation of consumption time in an affluent society. Metroeconomica 66(3):534–563.

271

Andreas Chai and Zakaria Babutsidze Fontana, R., & Guerzoni, M. (2008) Incentives and uncertainty: An empirical analysis of the impact of demand on innovation. Cambridge Journal of Economics 32:927–946. Foster, J., & Metcalfe, J. S. (2012) Economic emergence: An evolutionary economic perspective. Journal of Economic Behavior & Organization 82(2–3):420–432. Frank, R. (1985) The demand for unobservable and other nonpositional goods. American Economic Review 75(1):101–116. Frenzel Baudisch, A. (2007) Consumer heterogeneity evolving from social group dynamics: Latent class analyses of German footwear consumption 1980–1991. Journal of Business Research 60(8):836–847. Georgescu-Roegen, N. (1983) Herman Heinrich Gossen: His life and works in historical perspective, introductory essay to The Laws of Human Relations, by H. H. Gossen, Griskevicius, V., & Kenrick, D. (2013) Fundamental motives: How evolutionary needs influence consumer behavior. Journal of Consumer Psychology 23(3):372–386. Guerzoni, M. (2010) The impact of market size and users’ sophistication on innovation: The patterns of demand. Economics of Innovation and New Technology 19(1):113 Harper, D., & Endres, A. (2012) The anatomy of emergence, with a focus upon capital formation. Journal of Economic Behavior & Organization 82(2–3):352–367. Heffetz, O. (2011) A test of conspicuous consumption: Visibility and income elasticities. Review of Economics and Statistics 93(4):1101–1117. Hohnisch, M., Pittnauer, S., & Stauffer, D. (2008). A percolation-based model explaining delayed takeoff in new-product diffusion. Industrial and Corporate Change 17(5):1001–1017. Hopkins, Ed., & Kornienko, T. (2004) Running to keep in the same place: Consumer choice as a game of status. American Economic Review 94(4):1085–1107. Ida, T. (2010) Coevolution of product quality and consumer preferences. Journal of Bioeconomics 12(2):101–117. Israel, M. (2005) Services as experience goods: An empirical examination of consumer learning in automobile insurance. American Economic Review 95:1444–1463. Jackson, T., Jager, W., & Stagl, S. (2004) Beyond insatiability: Needs theory, consumption and sustainability. ESRC Sustainable Technologies Programme Working Paper Series 2:1–34. Janssen, M. A., & Jager, W. (2002) Stimulating diffusion of green products. Journal of Evolutionary Economics 12(3):283–306. Jeppsen, L., & Molin, M. (2003) Consumers as co-developers: Learning and innovation outside the firm. Technology Analysis and Strategic Management 15:363–383. Kaus, W. (2013) Beyond Engel’s law-A cross-country analysis. The Journal of Socio-Economics 47:118–134. Kiesling, E., Günther, M., Stummer, C., & Wakolbinger, L. M. (2012) Agent-based simulation of innovation diffusion: A review. Central European Journal of Operations Research 20(2):183–230. Kindleberger, C. (1989) Long waves in economics and politics. Economics & Politics 1:201–206. Kleinkrecht, A., & Verspagen, B. (1990) Demand and innovation: Schmookler re-examined. Research Policy 19:387–394. Kuznets, S. S. (1973) Modern economic growth: Findings and reflections. The American Economic Review 63(3):247–258. Lades, L. K. (2013) Explaining shapes of Engel curves: The impact of differential satiation dynamics on consumer behavior. Journal of Evolutionary Economics 23(5):1023–1045. Langlois, R. (2001) Knowledge, consumption, and endogenous growth. Journal of Evolutionary Economics 11:77–93. Lavoie, M. (1994) A Post Keynesian approach to consumer choice. Journal of Post Keynesian Economics 16(4):539–562. Lintott, J. (1998) Beyond the economics of more: The place of consumption in ecological economics. Ecological Economics 25:239–248. Lipsey, R. G., Carlaw, K. I., & Bekar, C. T. (2005) Economic Transformations: General Purpose Technologies and Long-Term Economic Growth: General Purpose Technologies and Long-Term Economic Growth. Oxford University Press. Lorentz, A., Ciarli, T., Savona, M., & Valente, M. (2016) The effect of demand-driven structural transformations on growth and technological change. Journal of Evolutionary Economics 26(1):219–246.

272

Evolutionary consumer theory Malerba, F., Nelson, R., Orsenigo, L., & Winter, S. (2007). Demand, innovation, and the dynamics of market structure: The role of experimental users and diverse preferences. Journal of Evolutionary Economics 17(4):371–399. Manig, C., & Moneta, A. (2014) More or better? Measuring quality versus quantity in food consumption. Journal of Bioeconomics 16:155–178. Markey-Towler, B. (2018) A formal psychological theory for evolutionary economics. Journal of Evolutionary Economics 28(4):691–725. Menger, C. (1871) Principles of Economics, Glencoe, Illinois, The Free Press. Metcalfe, S. (2001) Consumption, preferences, and the evolutionary agenda. Journal of Evolutionary Economics 11:37–58. Metcalfe, S., Foster, J., & Ramlogan, R. (2006) Adaptive economic growth. Cambridge Journal of Economics 30:7–32. Mowrey, D., & Rosenberg, N. (1979) The influence of market demand upon innovation: A critical review of some recent empirical studies. Research Policy 8:102–153. Moneta, A., & Chai, A. (2013). The evolution of Engel curves and its implications for structural change theory. Cambridge Journal of Economics, 38, 895–92310.1093/cje/bet033. Mueller, M., Schrempf, B., & Pyka, A. (2015) Simulating demand-side effects on innovation. International Journal of Computational Economics and Econometrics 5:220–236. Nelson, R., & Consoli, D. (2010) An evolutionary theory of household consumption behavior. Journal of Evolutionary Economics 20:665–687. Nelson, R., & Rosenberg, N. (1993) Technical Innovation and National Systems, in: R. Nelson (ed.) National Systems of Innovation. A Comparative Analysis, Oxford, Oxford University Press. Nelson, R., & Winter, S. (1982) An Evolutionary Theory of Economic Change, Cambridge Massachusetts, Harvard University Press. Nieman, B., & Vavra, J. (2019) The rise of niche consumption. NBER Working paper no. 26134. Nisticò, S. (2005) Consumption and time in economics: Prices and quantities in a temporary equilibrium perspective. Cambridge Journal of Economics 29:943–957. Pasinetti, L. L. (1981) Structural Change and Economic Growth, Cambridge, UK, Cambridge University Press. Rengs, B., & Scholz-Wäckerle, M. (2019) Consumption & class in evolutionary macroeconomics. Journal of Evolutionary Economics 29(1):229–263. Robson, A. J. (2001) The biological basis of economic behavior. Journal of economic literature 39(1):11–33. Røpke, I. (1999) The dynamics of willingness to consume. Ecological Economics 28:399–420. Ruprecht, W. (2005) The historical development of the consumption of sweeteners-a learning approach. Journal of Evolutionary Economics 15(3):247–272. Saad, G. (2007) The evolutionary bases of consumption, Psychology Press. Saad, G. (2013) Evolutionary consumption. Journal of Consumer Psychology 23(3):351–371. Safarzynska K., & van den Bergh J. C. J. M. (2010) Demand-supply coevolution with multiple increasing returns: Policy analysis for unlocking and system transitions. Technological Forecasting and Social Change 77(2):297–317. Saviotti, P.-P. (1996) Technological evolution, variety and the economy, Edward Elgar. Saviotti, P.-P. (2001) Variety, growth and demand. Journal of Evolutionary Economics 11:119–142. Saviotti, P. P., & Pyka, A. (2013) The co-evolution of innovation, demand and growth. Economics of Innovation and New technology 22(5):461–482. Scherer, F. M. (1982) Demand-pull and technological invention: Schmookler revisted. The Journal of Industrial Economics:225–237. Schlaile, M. P., Mueller, M., Schramm, M., & Pyka, A. (2018). Evolutionary economics, responsible innovation and demand: Making a case for the role of consumers. Philosophy of Management 17(1):7–39. Schmookler, J. (1960) Invention and Economic Growth, De Gruyter. Schumpeter, J. (1934) The theory of economic development, Transaction publishers. Shugan, S. (2006) Are consumers rational? Experimental Evidence? Marketing Science 25:1–7. Simon, H. (1956) Rational Choice and the Structure of the Environment. Psychological Review Steedman, I. (2001) Consumption takes time: implications for economic theory, Routledge.

273

Andreas Chai and Zakaria Babutsidze Swann, G. M. (2002) There’s More to Economics of Consumption than (almost) Unconstrained Utility Maximization, in A. McMeekin, K. Green, M. Tomlinson, & V. Walsh (eds.) Innovation by Demand, Manchester, Manchester University Press, pp. 23–40 Valente, M. (2012) Evolutionary demand: A model for boundedly rational consumers. Journal of Evolutionary Economics 22:1029–1080. Van den Ende, J., & Dolfsma, W. (2005). Technology-push, demand-pull and the shaping of technological paradigms-Patterns in the development of computing technology. Journal of Evolutionary Economics 15(1):83–99. Veblen, T. (1898). Why is economics not an evolutionary science? Quarterly Journal of Economics 12:373–397. Volland, B. (2012). The history of an inferior good: Beer consumption in Germany (No. 1219). Papers on Economics and Evolution. Von Hippel, E. (2005). Democratizing Innovation, MIT Press. Witt, U. (2017) The evolution of consumption and its welfare effects. Journal of Evolutionary Economics 27(2):273–293. Witt, U. (2001) Learning to consume – A theory of wants and the growth of demand. Journal of Evolutionary Economics 11(5):23–36. Witt, U. (2008) Propositions about novelty. Journal of Economic Behavior and Organization. Windrum, P. (2005) Heterogeneous preferences and new innovation cycles in mature industries: The amateur camera industry 1955–1974. Industrial and Corporate Change 14:1043–1074. Woersdorfer, J. S. (2010). When do social norms replace status‐seeking consumption? An application to the consumption of cleanliness. Metroeconomica 61(1):35–67. Zeppini, P., & Frenken, K. (2018) Networks, percolation, and consumer demand. Journal of Artificial Societies and Social Simulation 21(3):1.

274

21 EVOLUTIONARY PRICE THEORY Harry Bloch

Economic evolution involves structural changes from within, so evolutionary price theory needs to address how prices facilitate and accommodate these structural changes and how structural changes in turn impact prices. Such analysis is impossible using neoclassical price theory in which endowments of inputs, production technology and consumer preferences are all treated as exogenously determined.1 Basic elements of an alternative theory of price determination compatible with structural changes from within are outlined in this entry. Schumpeter provides essential preliminaries. In The Theory of Economic Development (Schumpeter, 1961 [1934]) and in Business Cycles (Schumpeter, 1939) the relationship between the stability of prices and innovation is used to explain why innovations ebb and flow, thereby generating cycles in the aggregate price level and alternating periods of stability and instability in relative prices. In Capitalism, Socialism and Democracy (Schumpeter, 1976 [1942]), innovation-driven structural changes come with monopolistic pricing practices, including price rigidity, with creative destruction providing the competition that counts in terms of rising living standards over long periods. Modern contributions to evolutionary price theory, beginning with Nelson and Winter (1982), identify ’Schumpeterian competition’ with the selection process accompanying the diffusion of innovations and show this process leads to price reductions along with a tendency towards increased market concentration. Metcalfe (1998) analyses the impact of the diffusion of innovations using replicator dynamics in which creative destruction leads to dominance of the fittest firms in line with Fisher’s Principle of evolutionary change. A parallel approach to analysing adjustment to disruption through innovation is provided by the theory of industry life cycles (Klepper, 1997). Of course, most markets are not usually disrupted. Nelson (2013) argues the need for a broader evolutionary approach to price determination to include markets not experiencing disruption through innovations. He suggests using the concept of market order to replace market equilibrium from neoclassical price theory, thereby adapting supply and demand analysis to be consistent with an evolutionary perspective. Bloch and Metcalfe (2018) argue market order is also generally observed in markets with dominant firms, where administrative routines replace market interactions of supply and demand in creating order. DOI: 10.4324/9780429398971-23

275

Harry Bloch

The interaction of disrupted and orderly markets generates the dynamics of prices for the economy as a whole, including business cycles associated with clusters of innovations as in Schumpeter (1939). The next section discusses the determination of prices in orderly markets. Following is a review of models of Schumpeterian competition with their implications for the movement of prices during the process of creative destruction. The penultimate section presents an analysis of price dynamics in the economy as a whole. Concluding is a summary of the role of prices in economic evolution along with observations on how the evolutionary approach differs from that of neoclassical theory.

21.1

Prices in orderly markets

Nelson (2013, p. 17) argues the need for an evolutionary price theory applicable to ‘the subjects dealt with in conventional price theory, that is how markets determine the configuration of prices and quantities that one observes at any time.’ Assumptions of optimisation and equilibrium are rejected as inconsistent with continuing structural change in an evolving economy. Nonetheless, most markets in an evolving economy are orderly at any point in time. According to Nelson (2013, p. 29), ‘An orderly market is characterized by a set of routines established over time that when employed by potential buyers and potential sellers are tuned to each other and generally result in transactions that are satisfactory for most parties on both sides of the market.’ He then suggests using supply and demand analysis based on behavioural routines, rather than optimisation, to analyse price determination in orderly markets with large numbers of both buyers and sellers. Order is created as the routines of buyers lead to increased purchases at lower prices and supplier routines result in offering to sell more at higher prices. Bloch and Metcalfe (2018) suggest extending application of the concept of market order to markets with dominant buyers or sellers. They note the widespread use of administrative routines by dominant firms to set prices fits well with evolutionary analysis, providing a reasoned response to the need for internal control to achieve strategic objectives in a complex environment. They then point to the large literature on administered prices in postKeynesian price theory, which can be adapted for evolutionary price theory dealing with markets having dominant buyers or sellers in parallel with how Nelson adapts conventional supply and demand theory for markets with large numbers of buyers and sellers. Importantly, the presumptions of post-Keynesian theory are consistent with those of evolutionary economics. Behaviour is purposeful, but characterised by rules and routines based on bounded rationality rather than optimisation. Rules and routines are used because current information is imperfect and the future is unknown. There is even scope for dominant firms to establish order in markets where demand curves in the usual sense don’t exist, such as with the introduction of radically new products. Further, the range of market outcomes deemed satisfactory by firms can include an amount of underutilised productive capacity. Market order thus extends beyond balancing supply and demand and certainly doesn’t depend on satisfying first-order conditions for maximising profits. Setting an administered product price involves adding an allowance for profit margin to an accounting measure of unit cost, which can be expressed as unit cost times a price-cost margin exceeding one (Lee, 1998). As the unit cost measure is generally based on historical cost or operating at what is considered normal output, changes in output have no immediate effect on prices. In contrast, changes in operating cost due to changes in input prices can be fully and quickly passed on to changes in product price. Thus, the pattern of administered price changes 276

Evolutionary price theory

in response to demand and cost changes is different than the pattern of price change under Nelson’s adapted supply and demand theory, where changes in demand and cost are symmetric in terms of being generally partially passed on to price changes.

21.2

Prices in motion

As Schumpeter (1976 [1950], p. 82) observes, ‘Capitalism, then, is by nature a form or method of economic change and not only never is but never can be stationary.’ A stationary economy would have all production and transactions recurring in identical fashion period after period. These are ideal conditions for entrepreneurs to reliably determine the prospective profits from innovations and obtain finance to acquire means of production for implementation. Thus, capitalist institutions that enable entrepreneurs to finance and profit from innovations ensure the economy can never remain stationary. Schumpeter distinguishes between inventions and innovations, with the invention of a new way of doing things only becoming an innovation when it is introduced into the economy.2 Implementing new ways of doing things generally meets resistance and requires a special type of leadership, which is provided by the entrepreneur. A primary task of the entrepreneur is acquiring control of means of production (labour, machines, buildings, and materials), most of which are already employed in established production elsewhere. Schumpeter (1961 [1934], 1939) first emphasises the role of banks in providing credit to entrepreneur for this purpose, but later (Schumpeter, 1976 [1950]) recognises the role of large businesses internally reallocating resources to innovative products and processes. The economic character of innovations under capitalism is reflected in the requirement that they are expected to be profitable. Regardless of what other motivations entrepreneurs may have, capitalism imposes profitability as an essential criterion for innovations to be successfully introduced and diffused through the economy. Banks, venture capitalists, and corporate boards require demonstration of potential profitability before financing innovations, while retained earnings out of realised profits are a primary source for financing subsequent growth. Successful innovations disrupt the prevailing structure of production and consumption, setting off price dynamics as part of the evolutionary process of creative destruction. These price dynamics differ from the tâtonnement process of neoclassical price theory or the adjustment of market prices toward natural prices in classical price theory. Price dynamics during creative destruction are irregular, path dependent, and long lasting. There is no return to the relative prices prevailing before the innovations. If and when order is restored to the economy, the structure of relative prices is consistent with the new structure of production and consumption rather than with the original structure. The irregular nature of price dynamics accompanying innovation and restructuring in the aggregate economy is emphasised in Business Cycles (Schumpeter, 1939). Innovations, especially when financed by newly created credit, lead to increased demand for the means of production, driving up costs for innovators and established firms alike. Increasing prices thus characterise the implementation phase of a cluster of innovations in the upswing of the business cycle. Later, outputs from the innovations compete with those from established producers, putting downward pressure on all prices. The downswing of the business cycle thus is characterised by generally falling prices. While Schumpeter suggests a typical sinusoidal pattern for price movements over business cycles, he recognises innovations are by nature uncertain and the reaction to them sets off secondary phenomena. Thus, the 277

Harry Bloch

magnitude and duration of each stage of the cycle is variable and there are also overlapping cycles of different lengths.3 Structural change associated with the diffusion of innovations implies differential rates of change of prices for different products within the movements in overall prices. Modern analysis of the process of structural change through innovation has been stimulated by Nelson and Winter (1982), who use computer simulation models to analyse evolution within industries. Firms search for innovations and previous innovations diffuse through the selection process of differential firm growth as well as through search efforts of noninnovating firms to imitate. Search efforts have stochastic outcomes changing the relative competitive positions of firms in the industry. In the Nelson and Winter analysis, successful innovators earn above-normal profits from their superior products or production processes and use these profits to expand capacity relative to that of rivals. They may also continue research and development activities towards further cost reductions. Sufficient expansion of output relative to demand growth puts downward pressure on prices across the industry. Profits of innovators and established producers both fall, with established producers increasingly constrained from further expansion and then threatened with closure when prices fall below their operating costs. Declining prices are thus part of the lengthy process of creative destruction within industries that leads to displacement of established producers by innovators. Metcalfe (1998) provides analytical results for a simplified model of the selection process within industries, starting with heterogeneity in characteristics across firms. Each firm maintains constant characteristics over time as there is no further search for innovations and imitations. Firm growth is proportional to each firm’s profits, with profit variation across firms due to the heterogeneity of firm characteristics. In the simplest model, low-cost firms are the fittest in terms of having the fastest growth. These firms increase market share, leading to a pattern of falling average unit cost following Fisher’s Principle that the rate of change in the average value (weighted by share of population) of a population characteristic is proportional to the weighted variance in that characteristic across the population. Average prices decrease at the same rate as average unit cost in Metcalfe’s analysis where prices are determined by imposing the requirement that the growth in capacity meets a constant rate of demand growth. This process takes time and slows once the fittest firms come to dominate the industry and the share-weighted variance of unit cost declines.4 The attractor value of the characteristic only fully dominates at the infinite horizon. Diffusion of innovations takes time in the analyses of Metcalfe (1998) and Nelson and Winter (1982) because expansion of productive capacity by innovating and imitating firms is limited by realised profits. Haas (2016) examines the pattern of diffusion of innovations when access to skilled labour or some other input in limited supply poses an additional limit on the growth of firms. If the labour supply limit is binding, innovating firms expand relative to firms using the old method by offering a higher wage and forcing the other firms to operate below productive capacity. Under the assumed conditions of labour rationing, investment in productive capacity, and market clearing for output, product price remains constant during the period in which an old method of production is displaced by a superior new method with lower labour requirements. Only after this period does further investment in the new technique lead to a declining output price.5 Markey-Towler (2016) proposes a general framework for analysing price dynamics during the diffusion of innovations, which includes the possibility that inadequate demand 278

Evolutionary price theory

imposes limits on the growth of firms during the diffusion period. Each firm faces demand dependent on the price and attributes of its product as well as on overall consumer demand. The firm’s strategic choices of pricing rule and product attributes then determine the limit that its own demand imposes on its expansion as underlying consumer demand expands. The demand limit determines the growth of the firm as long as the firm’s productive capacity is not binding. Relative growth of innovating firms then depends on strategic choices of all firms as well as on any cost or product attribute advantage of the innovating firm. Not surprisingly, nothing definite can be concluded about price dynamics without restrictions being imposed on the assumptions about firm behaviour and consumer responses. However, irregular price dynamics are a likely outcome as firms experiment with strategies in the ever-changing external environment.

21.3

Prices in the economy as a whole

Evolutionary price theory has a micro-meso-macro structure in line with other evolutionary economic theorising (Dopfer, 2005). At the micro level, individual firms have distinct product attributes, production processes and pricing rules. Heterogeneity is common in all these characteristics, even when firms compete for the same customers. At the meso level, average price for firms grouped into an industry of firms with similar products or production processes is influenced by differential growth of the heterogeneous firms as well as by adaptation of individual firm characteristics to the industry environment. Finally, there are common macro-level influences on all firms in the economy, including prices of primary inputs and aggregate incomes of consumers. Starting at the micro level, the product attributes and production processes of each firm determine an input-output relationship for its products. Unit variable cost corresponding to this relationship is the sum of unit cost for labour and unit cost of intermediate inputs. Unit cost may be calculated from historical cost accounting data or based on expected levels of inputs, outputs and input prices. Applying an administered pricing rule of the type discussed in the section above on orderly markets results in a price equation for each product as follows:

pij = mij

vij

(21.1)

where pij is product price for the jth firm in the ith industry, vij is the firm’s variable cost per unit and mij is the firm’s price-cost margin to cover overheads and profit, with mij > 1.0.6 Moving to the meso level, the weighted average price for firms grouped into an industry producing substitutable products is given by: pi =

j

(pij

sij )

(21.2)

where sij gives the jth firm’s share of the value of sales in the ith industry. In simple replicator models of diffusion of an innovation, such as Metcalfe (1998), the industry average price in (21.2) falls over time with the rising share of the fittest firms, those with the lowest unit cost or margins, even though each firm maintains a constant unit variable cost and a constant margin. This process of price reduction continues as long as there is differential growth favouring the fittest firms, although at a slackening pace with a falling share of industry sales from the high-price firms.7 279

Harry Bloch

Prices across the economy are linked together through common influences on variable cost. Each industry uses the products of other industries as intermediate inputs, so the price of each industry’s product is reflected in the cost of producing other products. Also, there are common influences on the wages paid by different industries. The influences of intermediate input prices and wages on unit variable cost are shown in

vij = wij

lij +

k

pijk

aijk

(21.3)

where lij is the amount of labour used in producing a unit of output by the jth firm in industry i and wij is the wage paid for that labour, while aijk is the amount of product k used in the production of one unit of output by firm j and pijk is the corresponding price.8 Substituting from (21.3) into (21.1) and then aggregating as in (21.2) gives an equation for the average industry price in terms of its proximate determinants as follows:

pi =

j

sij

mij

(wij

lij +

k

pijk

aijk )

(21.4)

Evolutionary influences on prices for the economy as a whole in (21.4) occur at the micro, meso, and macro levels. At the micro level, average prices are influenced by the strategic choice of price-cost margins by individual firms as well as their individual efforts to lower costs through increasing productivity in the use of labour and intermediate inputs and through driving down input prices. Average prices are influenced at the meso level as shares of the fittest firms increase due to selection processes associated with the diffusion of innovations. Finally, at the macro level, average price in each industry depends on the prices of products in other industries in proportion to the intensity of their use as intermediate inputs, while wages for the labour used in every industry are subject to common influences from the macro economy and the movement of labour between industries in response to wage differentials. The fact that prices in each industry depend on prices in the other industries in (21.4) complicates the analysis of price movements. Applying matrix algebra to the set of equations for all industries, yields the following solution for the vector of industry average product prices:

p = m [I

A] 1wl

(21.5)

where m is a vector of average price-cost margins weighted by sales shares of all firms in each industry, A is a matrix of input similarly weighted intermediate input requirements, w is a vector of similarly weighted wage rates, and l is the corresponding appropriately weighted vector of labour requirements. While market shares don’t appear explicitly in (21.5), they nonetheless impact on the average values for m, A, w, and l through the weighting given to individual firms in calculating the averages. In particular, the weights given to innovating firms rise with their relative expansion in the process of diffusing innovations, thereby changing average price at the industry level without any change in values at the firm level.9

21.4

Conclusion

Schumpeter (1961 [1934], 1939) argues innovation-driven growth leads to cycles in prices at the level of the economy as a whole. When markets are orderly, entrepreneurs are able to 280

Evolutionary price theory

reliably determine the profits of potential innovations and convince banks to provide finance. Credit created by banks allows the entrepreneurs to bid away means of production from established producers. Increases in wages of labour and the price-cost margins for producers of intermediate inputs such as industrial raw materials thus create an upswing in prices across the economy according to the relationship in (21.5). Production from innovators initially only has small impact on prices for the economy as a whole. The share of industry output coming from innovators starts small so they have little impact on average values of price-cost margins, intermediate input requirements, wage rates and labour requirements. Also, productivity or input price advantages of the innovators tend to be offset by increases in their price-cost margins that provide the basis for entrepreneurial profits. The impact of innovations on prices increases over time. Entrepreneurs use these profits to expand production capacity. They then increase their output shares through lowering price-cost margins to shift buyers from established producers, thereby generating a downswing in prices for the economy as a whole. Further indirect downward pressure on prices occurs when entrepreneurs repay their loans with interest to the banks, reducing outstanding credit and the money supply.10 When output from innovators and their imitators has substantially replaced that from established producers who fail to adapt, markets can again be orderly. The economy approaches a new normal with a fundamentally different structure. Firms, industries and occupations have been created as others have been destroyed. Productivity improvements reduce the constraints on achieving desired objectives, but scarcity persists in altered form. Prices have stabilised and the stage is set for a new wave of innovations. In neoclassical price theory, endowments of inputs, technology and preferences are all taken as exogenously determined and the economic problem is posed as one of scarcity. Equilibrium prices are determined by firms and households maximising their objectives subject to given constraints. In an evolving economy, scarcity is a challenge as well as a constraint. Schumpeter understood capitalism provides generous incentives to entrepreneurs and their backers to overcome the challenge of scarcity through innovation, with innovation becoming the driving force of economic development. The role of prices in fostering innovations and in facilitating their diffusion throughout the economy differs from the role of prices in efficiently dealing with scarcity. Incorporating mechanisms through which prices operate in these different roles is the key to developing an evolutionary price theory appropriate to understanding the dynamic forces behind economic development. Crucial is recognising that innovation generates variety, which leads to structural change through competition as a selection process. Schumpeter pointed us in the right direction to understanding the role of prices in a developing economy and more recent contributions have filled in parts of the picture. Still, the literature on evolutionary price theory is sparse and fragmentary. Much work remains.

Notes 1 The standard neoclassical assumptions of equilibrium and optimisation are unsuitable for evolutionary analysis because equilibrium is incompatible with development from within and optimisation is implausible when there are unforeseeable disruptions from innovations. 2 Also, innovation may not involve invention, ‘Although most innovations can be traced to some conquest in the realm of theoretical or practical knowledge that has occurred in the immediate or remote past, there are many which cannot’ ( Schumpeter, 1939, p. 84).

281

Harry Bloch 3 See Bloch (2018a, Chapter 3) for a detailed discussion of Schumpeter’s treatment of price dynamics over the business cycle. 4 When firms differ in other characteristics, such as product attributes or the propensity to reinvest profits, evolutionary processes are still at play but determining which firms are fittest and how fast prices decrease is more complicated. 5 Price dynamics differ under other assumptions about capital and labour intensity of old and new techniques, but generally the price dynamics are similar to those of Schumpeter’s business cycle discussed above. 6 Implicit in having each firm with distinct price, margin and variable cost is either the assumption that products are differentiated with distinct attributes, as in Markey-Towler (2016), or the assumption that buyers have imperfect price information, as in Almudi, et al. (2020). 7 If imitation by high-cost firms is successful in lowering their costs towards those of the fittest firms, the rate of decline in industry average price tends to temporarily accelerate. 8 For simplicity, it is assumed only one type of labour is employed by each firm. Multiple types of labour and corresponding wages could be included in (21.3) or the labour input requirement could be interpreted as an index multiplied by the corresponding average wage. Each firm is treated as paying a distinct price for its inputs of intermediate product and labour, recognising that access to cheap labour or intermediate input can provide a firm with an important competitive advantage. Differences in input prices across firms are subsumed into industry averages, but remain potential drivers of differential firm growth and the dynamics of industry average cost and price. 9 The matrix algebra structure of the system of equations in (21.5) is similar to those for the systems of equations used to solve for equilibrium prices in competitive general equilibrium and for reproduction prices in Sraffa’s (1960) analysis of production of commodities by means of commodities. However, neither of these other systems includes heterogeneity of values for individual production units, so industry prices remain constant without variation in individual firm input requirements, wage rates or profit margins. See Kurz (2008), Haas (2015) or Bloch (2018b) for further discussion of the relationship between evolutionary and Sraffian approaches to price determination. 10 Schumpeter (1939) adds speculation and overreaction to this basic two-phase cycle of upswing and downswing, resulting in a four-phase cycle of prosperity, recession, depression and recovery. He also adds overlapping cycles of different durations. Schumpeter’s theory of the money and credit is explained in detailed in the posthumously published, Treatise on Money ( Schumpeter, 2014), the effects of institutional changes in the financing of innovations since Schumpeter’s time are discussed in Callegari (2018).

References Almudi, Isabel, Francisco Fatas-Villafranca, Jesus Palacio and Julio Sanchez-Choliz (2020), ‘Pricing routines and industrial dynamics’, Journal of Evolutionary Economics, 30(3): 705–739. Bloch, Harry (2018a), Schumpeter’s Price Theory, London, Routledge. Bloch, Harry (2018b), ‘Neo-Schumpeterian price theory with Sraffian and post-Keynesian elements’, Journal of Evolutionary Economics, 28(5): 1035–1051. Bloch, Harry and Stan Metcalfe (2018), ‘Innovation, creative destruction and price theory’, Industrial and Corporate Change, 28(5): 1035–1051. Callegari, Beniamio (2018), ‘The finance/innovation nexus in Schumpeterian analysis: Theory and application in the case of U.S. trustified capitalism’, Journal of Evolutionary Economics, 28(5): 1175–1198. Dopfer, Kurt (2005), ‘Evolutionary economics: A theoretical framework’, in Kurt Dopfer, editor, The Evolutionary Foundations of Economics, Cambridge, Cambridge University Press, 3–55. Haas, David (2015), ‘Diffusion dynamics and creative destruction in a simple Classical model’, Metroeconomica, 66(4): 638–660. Haas, David (2016), ‘The evolutionary traverse: A causal analysis’, Journal of Evolutionary Economics, 26(5): 1173–1197. Klepper, Steven (1997), ‘Industry life cycles’, Industrial and Corporate Change, 6(1): 145–181. Kurz, Heinz D. (2008), ‘Innovations and profits: Schumpeter and the classical heritage’, Journal of Economic and Behavioral Organization, 67(1): 263–278.

282

Evolutionary price theory Lee, Frederic S. (1998), Post-Keynesian Price Theory, Cambridge, Cambridge University Press. Markey-Towler, Brendan (2016), ‘Law of the jungle: Firm survival and price dynamics in evolutionary markets’, Journal of Evolutionary Economics, 26(3): 655–696. Metcalfe, J. Stanley (1998), Evolutionary Economics and Creative Destruction, London, Routledge. Nelson, Richard R. (2013), ‘Demand, supply and their interaction on markets, as seen from the perspective of evolutionary economic theory’, Journal of Evolutionary Economics, 23 (1): 17–38. Nelson, Richard R. and Sidney G. Winter (1982), An Evolutionary Theory of Economic Change, Cambridge, MA, Harvard University Press. Schumpeter, Joseph A. (1961 [1934]), The Theory of Economic Development (translation of second German edition by Redvers Opie), London, Oxford University Press. Schumpeter, Joseph A. (1939), Business Cycles, New York, McGraw Hill. Schumpeter, Joseph A. (1976 [1950]), Capitalism, Socialism and Democracy, 3rd Edition, New York, Harper and Row. Schumpeter, Joseph A. (2014), Treatise on Money, English translation by Ruben Alvarado of Das Wesen des Geldes (edited from Schumpeter’s 1930 manuscript by Fritz Karl Mann, Gottingen: Vandenhoeck and Ruprecht), Aalten, Netherlands, Wordbridge Publishing. Sraffa, Piero (1960), Production of Commodities by Means of Commodities, Cambridge, Cambridge University Press.

283

22 THE COEVOLUTION OF INNOVATION AND DEMAND Pier Paolo Saviotti

22.1

Introduction

There is now a large consensus about the role which has been played by innovation in the growth process that occurred since the industrial revolution. However, innovation on its own could never have had an impact on growth if there had not been a demand for new goods and services. The emergence of such a demand required (i) a disposable income with which consumers could purchase these goods and services, (ii) a set of preferences which made these goods and services desirable and (iii) human capital that allowed consumers to use these goods and services. In these cases, it is both the average values and the distribution of disposable income, preferences and human capital that affect the relationship between innovation and demand. The priority of demand or of innovation and supply as factors affecting growth has been a frequent subject of discussion in the economics literature and in innovation studies. In economics, the general emphasis on Keynesian policies prevailing from the 1950s to the 1970s was followed by a shift to monetarism and to more supply-oriented policies from the 1980s onwards and by a recent revival of Keynesianism. In innovation studies, the demandpull, technology-push debate (Freeman, Soete, 1997; Mowery, Rosenberg, 1979) summarized and extended the work of Schmookler (1966) and Schumpeter (1911), who had stressed the role of demand and of supply respectively. The induced innovation hypothesis maintained that innovation tends to save the most expensive factor (Hicks, 1932) or the factor the share of which has increased (Kennedy, 1964; Von Weitzacker, 1966). Recently, growing attention has been paid to demand in models of economic growth, both orthodox (Murphy, Shleifer and Vishny 1989; Matsuyama, 2002; Foellmi and Zweimuller, 2006) and evolutionary (Bianchi, 1998; Aversi et al. 1999; Andersen, 2001, 2007; Metcalfe, 2001; Saviotti, 2001; Witt, 2001; Ciarli et al. 2010). A very recent paper by Nelson and Consoli (2010) sketches a broad outline of such a demand theory. In spite of all this interest, the relationship between innovation and demand did not consist uniquely in the relative ‘weight’ of supply and demand, or of innovation and demand, but included the mechanisms of their interaction and the time horizon in which this interaction was analyzed. The analysis of the relationship between supply and demand

284

DOI: 10.4324/9780429398971-24

The coevolution of innovation and demand

has often been carried out in a short-term perspective, for example by trying to predict the effect on prices of a fall in demand relative to supply. In a long-run perspective, the problem becomes to explain at least how a demand that included a limited range of goods and services can have been transformed into one comprising a much wider range and a much higher quality of goods and services. In other words, the interaction between innovation and demand must be studied in the context of the long-run process of structural change which occurred since the industrial revolution. The co-evolution of innovation and demand is the subject of this chapter.

22.2 22.2.1

Conceptual background

Co-evolution and economic development

The concept of co-evolution has been used in the innovation literature to analyze the coevolution of technologies and institutions. Technologies cannot develop in an institutional vacuum but need appropriate institutions (Nelson, 1994; Saviotti, 2005; Saviotti, Pyka, 2013; Saviotti Pyka 2012; Saviotti Pyka Jun 2016). Such institutions are required to regulate new technologies, establish intellectual property rights, create the required infrastructures, support the collective interests of new technology and of the corresponding industry, etc. Examples of such co-evolution are mass production in the U.S. car industry, the emergence of synthetic dyes in Germany (Murmann, 2003) and biotechnology in the USA (Nelson, 2008). A more general interpretation of the concept of co-evolution can be proposed at a system level. We can expect different types of interaction to be possible amongst system components. In an extreme case, different system component can be completely independent. Alternatively, a component C1 could affect positively another component C2, while an increase in C1 would not have any effect on C2. Example of unidirectional effect, in which C1 would be the cause of changes in C2 but C1 could not affect C1. Co-evolution is a particular type of interaction that exists when two different components, C1 and C2, of a given system interact in such a way that changes in one of them, say C1, affect C2 and that changes in C2 affect C1. In general, innovations tend to emerge in a pre-institutional form, without institutions specific to the new technology. We can expect the development of new technology to be very limited unless the initial innovation is followed by the creation of appropriate institutions and infrastructures, which would provide the rules and the physical means required for the use of the new technology to be used for the advantage of society. A clear example is given by cars and roads: the market for cars has been considerably enhanced by the construction of roads and by rules that constrain their use. Thus, the more a new technology develops, the more the appropriate institutions need to grow giving rise to a feedback loop which would slow down only when the market(s) for the new technology were completely saturated. As compared to a case in which technologies and institutions are independent, coevolution can accelerate the rate of growth of the market for new technology before saturation is attained, but also lead to a more rapid decline when one of the interacting components is reduced. The type of co-evolution we are going to be concerned with in this paper is between innovation and demand. The analysis here will be mainly based on the TEVECON model of development by the creation of new sectors (Saviotti, Pyka, 2004, 2008, 2013; Research Gate). This co-evolutionary process occurs because sectoral search activities, which increase with sectoral demand, affect output price, quality and differentiation, which in turn affect

285

Pier Paolo Saviotti

demand. A positive feedback loop can be established which can give rise to a faster growth of both demand and innovation than it would have been possible if the two variables had not been influencing each other. In this sense, co-evolution works as autocatalysis (Nicolis and Prigogine, 1989) by using the output of one stage of the process as the input of the following stage.

22.2.2

Innovation and economic development

Most of the objects citizens of developed countries use in their everyday life did not exist before the industrial revolution, and many of them did before he beginning of the 20th century. The structural change that created all these innovations can be described by three stylized facts, that represent the three following long run trajectories: The efficiency of productive processes increases in the course of time. The net number of distinguishable industrial sectors resulting from the emergence of new sectors and from the extinction of pre-existing ones increases in the course of time. The output quality and internal diversification of most sectors increase in the course of time. Trajectories T2 and T3 mean that the output variety (T2) and the diversification of the economic system (T3) increase in the course of economic development. Examples of T2 are the emergence of cars, computers, television, portable phones, no analogue of which existed before the 20th century. These innovations did not substitute pre-existing goods and services but coexisted with them. Furthermore, the industrial sectors producing old and new goods and services became increasingly diversified and produced output of higher quality (T3). Thus, the range of product models of even large producers of cars, tractors, or washing machines is much wider today than it was in the 1950s. The existence of these trajectories has been predicted by evolutionary growth models (Saviotti Pyka, 2004, 2008, 2013; Ciarli et al. 2010) and confirmed by empirical work (Funke and Ruhwedel, 2001a, 2001b; Frenken et al. 2007; Boschma, 2017; Hidalgo et al. 2007, 2009; Felipe 2012; Freire, 2017). The T1, T2 and T3 trajectories were present in economic development but they were not independent. An economic system producing a constant set of goods and services with increasing efficiency could not develop due to the bottleneck created by the imbalance between increasing productive efficiency and saturating demand (Pasinetti, 1981, 1993). Such imbalance could have allowed to produce a constant output with a falling labour intensity, possibly leading to unemployment. Such a bottleneck could have been avoided by the emergence of sectors producing new goods and services. However, while increasing efficiency was created by some form of innovation, it was not the same type of innovation that could give rise to new goods and services. Here creativity is defined as the capability to generate entities qualitatively different from anything that existed before. For example, cars, computers, airplanes, television or portable phones could not have been generated by the increasing efficiency in the production of the goods and services that existed before them. Thus, the trajectories T1, T2 and T3 needed both efficiency and creativity. Trajectories T2 and T3, which involved the emergence of qualitatively new goods and services, could not have been explained only in terms of efficiency. 286

The coevolution of innovation and demand

Although the emergence of new sectors and their diversification required both efficiency and creativity, they were not independent but interacted in the process of economic development. Efficiency is here defined as the ratio of the outputs produced to the inputs used in a given production process, subject to the condition that the output produced is qualitatively constant. Thus, an increase in this output input (Q/I) ratio represents an increase in efficiency only if the output produced is qualitatively unchanged. For example, an increase in the Q/I ratio in the production of a given type of shoes can be interpreted as an increase in efficiency only if the shoes are qualitatively unchanged. If we were to measure an increase in the Q/I ratio while the shoes produced changed from a very basic to a high-quality type, the increase in Q/I could be due to a combination of a change in efficiency and of a change in creativity. Neither increasing efficiency nor creativity alone can explain the process of economic development. Creativity is today generated by search activities, a general analogue R&D (Nelson, Winter, 1982) that requires resources. Such resources are created by the surplus obtained due to the growing efficiency in the production of preexisting goods and services. Thus, efficiency and creativity are complementary forces jointly contributing to economic development. The above trajectories involve both innovation and demand. The creation of new sectors producing new goods and services requires innovation, investment, purchasing power and new preferences. The purchasing power required to purchase new goods and services arises from the combination of three mechanisms: (i) the surplus accumulated by growing efficiency in the production of pre-existing ones, (ii) the wages paid to workers employed in the production of new goods and services and (iii) demand-oriented economic policies.

22.2.3

The evolution of demand since the Industrial Revolution

At the beginning of the 19th century in the United Kingdom, the birthplace of the Industrial Revolution and at that time the richest country in the world, working-class households spent a very large share of their income on food, clothing and housing (Hobsbawm, 1968), with very little left for other types of consumption. The previous three items can be considered biological necessities, required for human survival. With very limited improvements this pattern of consumption remained almost the same throughout all of the 19th century and began to change only during the 20th century (Ibid). Today in post-industrial societies, most people consume a very wide range of goods and services. This change in consumption patterns involved changes in both supply and demand. Innovation increased productive efficiency for existing goods and services and created new ones. In general, the new goods and services did not substitute the older ones but were added to them thus leading to an increasing consumption variety. Why did this delay in output variety (T2) and intra-sector (T3) diversification occur only in the XXth century in spite of the fast growth in productive efficiency (T1) during the first phase of the industrial revolution? There are essentially two reasons why efficiency growth (T1) preceded T2 and T3. First, throughout human history most people barely managed to satisfy their primary needs, such as food, clothing and housing, that are an incompressible part of consumption. Production efficiency was just enough to supply most people with the basic necessities required for survival and they lacked the resources to satisfy other human wants. During the 19th century, the high rate of population growth absorbed all the rise in productive efficiency generated by the Industrial Revolution. In the 20th century, a number of factors, including growing levels of education, a slowdown in the rate of population 287

Pier Paolo Saviotti

growth and redistributive economic policies, contributed to create the disposable income required to go beyond basic needs. These changes in the composition of consumption suggested that human needs and wants were hierarchically organized and that consumers could start consuming other goods and services when their income per capita exceeded the one required to satisfy their basic needs. A hierarchical theory of wants has been discussed by leading economists at the end of the 19th century such as Léon Walras (1896, 1988), William Stanley Jevons (1924), Alfred Marshall (1949) and has received its most explicit and detailed treatment in the work of Carl Menger (1950). According to this theory wants can be ranked in order of absolute importance, with most basic wants at the bottom of the list (lower wants) and with the most sophisticated (higher wants) at the top (Maslow, 1943). Such hierarchical theory can be correct for what concerns the relationship between primary needs and higher wants, but it is much less useful for the ranking of higher wants. However, here it suffices to state that higher wants are likely to be satisfied only after a threshold represented by the minimum efficiency required to cover the primary needs of the whole population can be overcome. The trend towards an increasing diversification of consumption was anticipated by Veblen (1899), who observed that the rich did not consume mainly to satisfy their biological necessities but to present themselves in society and to establish their status. During the 20th century, the pattern of consumption of most people in industrialized countries became increasingly diversified and gradually included goods and services of growing quality. The previous considerations raise the question of the origin of the disposable income required to purchase the new goods and services generated by innovation. A partial answer is that such disposable income came from the surplus generated by the growing efficiency in the production of pre-existing goods and services. In addition to such surplus, the creativity needed to create new goods and services and adequate consumer preferences were required. It is unlikely that such preferences could exist for objects of consumption that were completely unknown to consumers. It is more likely that preferences were gradually formed by interaction between producers and users. As Schumpeter thought (Saviotti, 2001) producers had to educate consumers about new innovations. Furthermore, the efficiency generated surplus was only a necessary but not sufficient condition for the formation of a demand for innovations. Other conditions were the higher wages enabling workers to purchase the new goods and services, the higher competencies required to produce them and leading to new employment.

22.3

Coevolution and economic development by the creation of new sectors

TEVECON1 is both an endogenous growth model and a model in which the composition of the economic system, defined as the list of actors, activities and objects required to describe the system, changes endogenously in the course of time. The structure of an economic system, a concept related to composition, is defined as the set of components of the system and of their interactions. In TEVECON the components are industrial sectors, producing physical goods or services and other activities, such as education. The type of economic development analyzed by the model is the one described by the previous three trajectories: T1, T2 and T3. In TEVECON each sector is created by an important innovation, typically radical, qualitatively different from any pre-existing ones. Each innovation is introduced by a Schumpeterian entrepreneur acting in the expectation of a temporary monopoly. If the innovation is successful, it is followed by a bandwagon of imitators gradually raising the 288

The coevolution of innovation and demand

intensity of competition until it reaches the level prevailing in the mature sectors, corresponding to what Schumpeter called the circular flow. After their creation, the new sectors follow a life cycle in which the number of firms increases up to a maximum and then falls to a much lower value. The shape and duration of this life cycle depends on financial resources, the intensity of competition, failures, mergers and acquisitions, demand and the size of the corresponding markets. The inducement to the creation of new sectors comes from the saturation of demand in pre-existing ones and from the growing intensity of competition. We could call this situation, which corresponds to the passage of a sector from an innovative stage to the circular flow, generalized saturation.2 Entrepreneurs would then stop investing in saturated sectors and start investing in innovations which could still be at the stage of niches, however promising ones. Of course, this further investment could only occur if there is a pool of inventions ready to be exploited. The specification of demand in TEVECON is particularly important. In TEVECON, the demand for the output of a given sector i (Eq 22.1) depends positively on disposable income (DDisp,i), on product quality (Yi) and on product diversification (ΔYi) and negatively on product price (pi). Furthermore, demand is affected by consumer preferences (kpref,i).

Dit = kpref , i Di0 DDisp, i

Yi

Yi

(22.1)

Pi

This demand function is somewhat unusual for the presence of output quality and output diversification. Here disposable income is generated by the surplus due to the growing productive efficiency in sectors older than i and by the income of workers in sector i. Further contributions to the disposable income of sector i could come from activities complementary to i, such as education. The presence of ΔYi in Eq 22.1 implies that demand for the output of i depends on the range of output quality offered. The coevolution of demand and innovation occurs because Di, product quality, product diversification and product price depend on search activities SEi, a general analogue of R&D (Eqs 22.2–22.4).

SEit = 1 + k4 1 Yit = yi0 +

exp

1 + exp[k14

Yit = yi 0 +

t 1 k5 Dacc ,i

1 k15 (SEit

1 + Exp [k16

(22.2)

SE 0 )]

1 k17 (SEit

SE 0 )]

(22.3)

(22.4)

Thus, search activities increase with demand, leading to a rise in product quality and product diversification, which in turn leads to further increase in demand. The nature of coevolution is demonstrated in an experiment in which two scenarios, called low quality (LQ) and high quality (HQ) were compared (Saviotti, Pyka, 2013a, 2013b). In LQ output diversification, represented by trajectory T3, was absent. In this scenario, a new sector continued to produce the same unchanged output throughout the sector life cycle. In the HQ scenario output quality and differentiation continued to increase during the sector life cycle. The two scenarios were compared for their capacity to create income, employment, demand and human capital (Figures 22.1–22.6). 289

Pier Paolo Saviotti 2y 200 150 100 50 – 1

251

Figure 22.1

501

751

1001

1251

1501

1751

t

Effect of product quality on aggregate income (Y). Light curve for the LQ scenario, thick curve for HQ scenario. Time on the horizontal axis.

employment

2 000 1 500 1 000 500 – 1

251

501

751

1001 1251 1501 1751

t

Figure 22.2 Effect of product quality on aggregate employment. Light curve LQ, thick curve HQ, Time on the horizontal axis. 1.2 1.0 0.8 0.6 0.4 0.2 – 1

Figure 22.3

101

201

301

401

Product quality (yit) in the low-quality (dark) and high-quality (light) case. Yi on the vertical axis, time on the horizontal axis.

The two scenarios lead to important macroeconomic differences: the income created by the HQ scenario is initially lower than that of LQ, but overtakes it after a given time (Figure 22.1). On the other hand, the rate of creation of employment is systematically and substantially higher in the LQ scenario (Figure 22.2). We can further observe that the small bumps in Figures 22.1 and 22.2 are shorter and more numerous in the LQ scenario in the HQ scenario. Such bumps, corresponding to sector life cycles, indicate that sectoral life cycles in the LQ scenario are shorter than those in the HQ scenario. This shorter duration of 290

The coevolution of innovation and demand 1.0

0.8

0.5

0.3

– 1

101

Figure 22.4

201

301

401

Sectorial demand (Dt i) in the low-quality (dark) and high-quality (light) case. Di on the vertical axis, time on the horizontal axis.

2.0 1.8 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 – 1

101

201

301

401

501

601

701

801

901

Figure 22.5 Sectoral output (qti) in the low-quality (dark) and high-quality (light) case. Qi on the vertical axis, time on the horizontal axis. 0.6 0.5 0.4 0.3 0.2 0.1 – 1

101

Figure 22.6

201

301

401

501

601

701

801

901

Quantity of human capital (HCi) used in a sector in the low-quality (dark) and highquality (light) case. HCi on the vertical axis, time on the horizontal axis.

291

Pier Paolo Saviotti

the LQ life cycle is completely logical because in this case only volume saturation of the market can occur, when the number of units of output produced in a given sector remains constant over time. On the other hand, in the HQ case, the value of the corresponding market can keep growing even after the market has become saturated in volume due to the increasing quality and diversification of the output of the sector. The previous macroeconomic differences can be explained by sectoral results. Sectoral output quality, sectoral demand (Figure 22.4), sectoral output (Figure 22.5) and sectoral wages (Figure 22.6) are initially lower in the HQ case, but then grow much faster than in the LQ case. These results seem to indicate that the economic development that took place since the Industrial Revolution is a combination of scenarios LQ and HQ, the former dominating in the initial period and the latter overtaking it later. This switch from LQ to HQ could not occur until productive efficiency reached the level required to supply biological necessities, as shown, for example, by the changing share of expenditure in working class household in the United Kingdom during the period 1840–1940 (Hobsbawm, 1968). The transition from the LQ to the HQ scenario was called by Saviotti and Pyka (2013b) ‘from necessities to imaginary worlds’, and corresponds to the consumption patterns observed during the 20th century in industrialized countries. Towards the end of the 19th century, Veblen (1899) observed that consumption of rich people was not intended mainly to satisfy biological needs but to display their social status. During the XXth century, and particularly in the second half of it, the consumption of most people in industrialized countries moved beyond basic needs by incorporating new types of goods and services and by purchasing higher quality and a greater variety of them. In this process, the shape of the demand curve changed systematically by including a part in which demand increases with price (Figure 22.7) (Saviotti, Pyka, 2017). We have so far shown that there are important interdependencies between demand and innovation (Eqs 22.1–22.4), that these interdependencies have great macro-economic and meso-economic implications (Figures 22.1–22.6). We studied these implications by exploring the two scenarios, LQ and HQ, that differ for the presence (HQ) or absence (LQ) of trajectory T3. In other words, in the LQ scenario new sectors are created but they do not diversify P 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 -

Figure 22.7

0.5

1.0

D

Effect of changing product quality and product diversification on the shape of demand curves. Product quality and product differentiation increase from the downward sloping curves to those having the steepest upward sloping part. Demand on the vertical axis, time on the horizontal axis.

292

The coevolution of innovation and demand

internally whereas they do so in the HQ scenario. As we previously pointed out, both T2 and T3 avoid the development bottleneck that would have occurred if increasing efficiency alone (T1) had been driving economic development, but they do it in quite different ways. Thus, different mixtures of T2 and T3 are possible and lead to different paths of economic development. In the LQ scenario output variety would have increased faster but wages, demand and human capital would have remained stagnant. On the other hand, in HQ wages, demand and human capital increase substantially more but employment and output variety rise more slowly, leading to longer industry life cycles. It seems as if the pattern of economic development observed in industrialized countries can be explained by a transition from LQ to HQ occurring at about the beginning of the 20th century (Saviotti, Pyka, 2013) In order for the LQ HQ transition to happen, the competencies and the purchasing power required to make and purchase the higher quality and more diversified goods and services must be generated. The generation of the higher competencies required an expanded education system while the purchasing power was generated by higher wages and by the growing employment in sectors producing higher quality and more diversified goods and services and by other activities producing complementary inputs, such as education (Figure 22.8). So far in this chapter, consumers were considered a homogeneous group. This is far from the reality prevailing in any society. Even in the most egalitarian society we can expect to find a distribution of income, competencies and preferences that will inevitably lead to a distribution of demand. For example, in modern societies, education is a likely codeterminant of the distribution of income and of demand. A similar situation has been explored by Saviotti and Pyka (2017) in a version of the TEVECON model containing two social classes, called L and H, differing for their levels of education and human capital, giving rise to a distribution of competencies and to a corresponding income distribution. The L and H classes can be compared to blue- and white-collar Falling rate of population growth

Increasing search Activities (SE) Increasing Quality & Diff

Increasing education

Higher product price

Disposable income, demand

Increasing Competencies Higher wages

No impact of Innovation on growth unless Innovative goods and services had not been purchased by consumers®co-evolution

Figure 22.8 Co-evolutionary mechanism partly explaining the advent of outputs of higher quality and differentiation in the process of capitalist economic development.

293

Pier Paolo Saviotti

workers, or to low- and high-skill workers. The shape of Engel curves differs considerably for the L and H classes. For example, under a wide range of circumstances, the Engel curves of the L class show oversaturation3 while those of the H class show non-saturation. Thus, demand saturation for the whole population of an economic system could be an artificial result of the aggregation of individual demands, none of which saturates.

22.4 Summary and conclusions In this chapter, coevolution was discussed as a mode of interaction between innovation and demand. Although the relative importance of demand and supply or of innovation and demand have been discussed very often in the economics and innovation literatures, they have always been discussed in terms of the relative weights of one or the other rather than of their mode of interaction. The assumption in this chapter is that not only both innovation and demand are important, but that their mode of interaction is important as well. Coevolution involves positive feedback between innovation and demand rather than their independent action or negative feedback. Innovation contributes by creating new sectors and diversifying them, while demand contributes by providing the resources required to create and diversify new sectors. Underlying these processes there is the interaction between efficiency and creativity. The coevolution of innovation and demand is one of the many possible types of coevolution. Other important examples are the coevolution of technologies and institutions or that of education, innovation and demand. All the variables involved in coevolution can in principle be expected to interact with other variables, as it would happen in a complex socioeconomic system. Thus, coevolution is just a special case of complexity.

Notes 1 TEVECON description: https://www.researchgate.net/publication/292130135 TEVECON Description of Model, DOI: 10.13140/RG.2.1.1626.9841 2 Generalized saturation can include aspects other than demand and competition, such as the rate of profit. It could even be dependent on conditions affecting market size, such as competing or complementary technologies, public policies, recessions or wars. 3 Oversaturation is here defined as situation in which Engel curves show a rise up to a maximum followed by a decline.

References Andersen, E.S. 2001. “Satiation in an Evolutionary Model of Structural Economic Dynamics.” In Escaping Satiation, the Demand Side of Economic Growth, edited by U. Witt, 165–186. Berlin: Springer. Andersen, E.S. 2007. “Innovation and Demand.” In The Elgar Companion to Neo-Schumpeterian Economics, edited by Y.H. Hanusch and A. Pyka, 754–765. Cheltenham: Elgar.Aversi et al. 1999. Aversi, R., Dosi, G., Fagiolo, G., Meacci, M., and Olivetti, C. 1999. “Demand Dynamics with Socially Evolving Preferences.” Industrial and Corporate Change 8: 353–399. Bianchi, M. ed. 1998. The Active Consumer. London: Routledge. Boschma R. 2017. “Relatedness as Driver of Regional Diversification: A Research Agenda.” Regional Studies 51(3): 351–364, March. Chai, A., and Moneta, A. 2010. “Retrospectives Engel Curves.” Journal of Economic Perspectives 24(1): 225–240. Ciarli, T., Lorentz, A., Savona, M., and Valente, M. 2010. “The Effect of Consumption and Production Structure on Growth and Distribution. A Micro to Macro Model.” Metroeconomica 61(1): 180–218.

294

The coevolution of innovation and demand Felipe, J., Kumar, U., Abdon, A., and Bacate, M. 2012. “Product Complexity and Economic Development”, Structural Change and Economic Dynamics 23(2012): 36–68. Foellmi, R., and Zweimuller, J. 2006. “Income Distribution and Demand Induced Innovations.” Review of Economic Studies 82: 95–112. Freeman, C., and Soete, L. 1997. The Economics of Industrial Innovation. London: Pinter. Freire, C. 2017 “Economic Diversification: Explaining the pattern of diversification in the global economy and its implications for fostering diversification in poorer countries”, UNU MERIT Working paper, #2017-033, ISSN 1871-9872. Frenken, K., van Oort, F.G., and Verburg, T. 2007. “Related Variety, Unrelated Variety and Regional Economic Growth.” Reg Stud 41(5): 685–697. Funke, M., and Ruhwedel, R. 2001a. “Product Variety and Economic Growth: Empirical Evidence for the OECD Countries.” IMF Staff papers 48(2). Funke, M., and Ruhwedel, R. 2001b. “Export Variety and Export Performance: Empirical Evidence from East Asia.” JAsian Econ 12: 493–505. Gualerzi, D. 2001. Consumption and Growth, Recovery and Structural Change in the US Economy. Cheltenham: Edward Elgar. Hicks, J.R. 1932. The Theory of Wages. London: MacMillan. Hidalgo, C.A., Klinger, B., Barabasi, A-L, and Hausmann, R. 2007. “The Product Space Conditions the Development of Nations.” Science 317(5837): 482–487. Hidalgo. C.A., and Hausmann, R. June 30, 2009. “The Building Blocks of Economic Complexity.” PNAS 106(26): 10575. Hobsbawm, E. 1968. Industry and Empire. Harmondsworth: Penguin Books. Jevons, W.S. 1924. The Theory of Political Economy, 4th edn. London: MacMillan. Kaldor, N. 1957. “A Model of Economic Growth.” The Economic Journal 67(268) (Dec., 1957): 591–624. Kennedy, C. 1964. “Induced bias in innovation and the theory of distribution.” Economic Journal 74 (1964): 541–547. Marshall, A. 1949. Principles of Economics. 8th ed. Macmillan. Maslow, A. 1943. “A Theory of Human Motivation.” Psychological Review, no 50, 1943, p. 370–396. Matsuyama, K. 2002. “The Rise of Mass Consumption Societies.” Journal of Political Economy 110: 1035–1070. Menger, C. 1871. Principles of Economics. Translated by J. Dingwall and B.F. Hoselitz, with an Introduction by Friedrich A. Hayek. New York University Press, 1981. Menger C. (1950). Principles of Economics, The Free Press, Glencoe, Illinois, Original edition Untersuchungen über die Methode der Socialwissenschaften, und der Politischen Oekonomie insbesondere, (1871). Metcalfe, J.S. 2001. “Consumption, Preferences and the Evolutionary Agenda.” Journal of Evolutionary Economics 11: 37–58. Michael Funke & Ralf Ruhwedel, 2001. Product Variety and Economic Growth: Empirical Evidence for the OECD Countries, IMF Staff Papers, Palgrave Macmillan, vol. 48(2), 1–1. https://ideas.repec.org/ s/pal/imfstp.html Mowery, D., and Rosenberg, N. 1979. “The Influence of Market Demand Upon Innovation: A Critical Review of Some Recent Empirical Studies.” Research Policy 8: 102–153. Murmann J.P. 2003. Knowledge and Competitive Advantage: The Co-evolution of Firms, Technologies and National institutions. Cambridge: Cambridge University Press. Murphy, K., Schleifer, A., and Vishny, R. 1989. “Industrialization and the Big Push.” Journal of Political Economy 97: 1003–1026. Nelson, R.R., and Consoli, D. 2010. “An Evolutionary Theory of Household Consumption Behaviour.” Journal of Evolutionary Economics 20: 665–687. Nelson, R. Winter, S., An Evolutionary Theory of Economic Change, Cambridge, Mass, Harvard University Press, (1982). Nelson, R.R. 1994. “The Co-evolution of Technology, Industrial Structure, and Supporting Institutions.” Industrial and Corporate Change 3(1): 47–63. Nelson R.R., What enables rapid economic progress: What are the needed institutions? Research Policy 37 (2008) 1–11. Nicolis, G., and Prigogine, I. 1989. Exploring Complexity. New York: Freeman. Pasinetti, L.L. 1981. Structural Change and Economic Growth. Cambridge University Press.

295

Pier Paolo Saviotti Pasinetti, L.L. 1993. Structural Economic Dynamics. Cambridge University Press. Research Gate, https://www.researchgate.net/publication/292130135 TEVECON Description of Model, DOI: 10.13140/RG.2.1.1626.9841. Saviotti, P.P. 2001. “Variety, Growth and Demand.” Journal of Evolutionary Economics 11: 119–142. Saviotti, P.P., and Pyka, A. 2017. “Innovation, Structural Change and Demand Evolution: Does Demand Saturate?” Journal of Evolutionary Economics 27: 337–358 DOI 10.1007/s00191-015-0428-2. Saviotti, P.P., and Pyka, A. 2012. “On The Co-Evolution Of Innovation And Demand: Some Policy Implications.” Revue de l’OFCE, n°124, pp 349-388 - Debates and policies: Agent-based models and economic policy, edited by Jean-Luc Gaffard and Mauro Napoletano. Saviotti, P.P., and Pyka, A. 2013a. “From Necessities to Imaginary Worlds: Structural Change, Product Quality and Economic Development.” Technological Forecasting & Social Change 80(2013): 1499–1512. Saviotti, P.P., and Pyka, A. 2013b. “The Co-evolution of Innovation, Demand and Growth.” Economics of Innovation and New Technology, DOI: 10.1080/10438599.2013.768492. Saviotti, P.P., Pyka, A., and Bogang Jun, B. 2016. “Education, Structural Change and Economic Development.” Structural Change and Economic Dynamics 38: 55–68. Saviotti, P.P. 2005. “On the co-evolution of technologies and Institutions.” In Towards Environmental Innovation Systems. edited by M. Weber, and J. Hemmelskamp, Berlin, Heidelberg, New York: Springer. Saviotti, P.P., and Pyka, A. 2004. “Economic Development by the Creation of New Sectors.” Journal of Evolutionary Economics 14(1) (2004): 1–35. Saviotti, P.P., and Pyka, A. 2008. “Micro and Macro Dynamics: Industry Life Cycles, Inter-sector Coordination and Aggregate Growth.” Journal of Evolutionary Economics 18(2008): 167–182. Saviotti, P.P., and Pyka, A. 2013. “From Necessities to Imaginary Worlds: Structural Change, Product Quality and Economic Development.” Technological Forecasting & Social Change 80: 1499–1512. Schmookler, J. 1966. Invention and Economic Growth. Cambridge, Mass: Harvard University Press. Schumpeter, J. 1911. The Theory of Economic Development. Cambridge, Mass: Harvard University Press (1934, original edition 1911). Veblen, T. 1899. The Theory of the Leisure Class: An Economic Study in the Evolution of Institutions. Macmillan. Walras L. (1896) Eléments d’économie politique Pure, 3rd edn. Economica, Paris (reprinted 1988). Witt, U. 2001. “Learning to Consume: A Theory of Wants and the Growth of Demand.” Journal of Evolutionary Economics 11: 23–36.

296

PART II

Evolutionary economic policy and political economy

23 EVOLUTIONARY ECONOMIC POLICY AND COMPETITIVENESS Michael Peneder

23.1

Introduction

This note elaborates on the concept of competitiveness as a comprehensive heuristic of evolutionary economic policy. The core objective is to replace the conventional emphasis on allocative efficiency and the associated rationalities of failure with an evolutionary focus on the drivers of economic development. Following Schumpeter (1911), one can define development by the growth in real income combined with qualitative transformations of the socio-economic system. Thus, we further define that competitiveness is the ability of an economic system to achieve high real incomes together with qualitative change, be it within firms, at the level of firm populations, sectors, or the aggregate economy, in a sustainable manner and in support of the overall goals of society. In other words, competitiveness is the ability to evolve in accordance with a long-term improvement of social and environmental living conditions. Competitiveness policy is the set of public interventions that aim to foster a system’s capacity to evolve in the above sense.1 To begin with, one must acknowledge that the common paradigm of ‘market failure’ has been extremely successful in rationalizing public interventions that we generally consider good and necessary, e.g. with regard to public goods, external effects, asymmetric infor­ mation, or indivisibilities that constrain competition. Its success originates in the smart choice of assumptions and rules that enabled a winning balance between flexibility in response to actual policy needs and comprehensible principles for general guidance. But as Nelson (2009, p. 9) points out, ‘these concepts and maxims are not logically tied to a structure of modern neoclassical economic theory. They are perfectly at home within an economic analysis structured by evolutionary theory.’ What difference does it then make, whether we apply one or the other logic of inter­ vention? Nelson highlights the more realistic approach to institutional complexity as one of the main advantages of evolutionary theorizing. Obviously, this is particularly important to the design and practice of economic policy. It specifically helps to overcome the privileged standing of pure market organization as the presumed default structure, which is misleading since markets are never perfect and never exist in a pure form. Instead, they are always conditioned by non-market institutions, which can be either more obstructive or supportive

DOI: 10.4324/9780429398971-26

299

Michael Peneder

of particular activities and outcomes. Accepting the high degree of uncertainty together with a strong dependency of the effectiveness of public interventions on the situational context renders the idea of optimized policy prescriptions untenable. Instead, the evolutionary approach emphasizes the importance of policy experiments and the need to provide insti­ tutional space for processes of discovery (Metcalfe, 1994; Cantner and Pyka, 2001). The present analysis adds further aspects to an evolutionary paradigm of economic policy. First and foremost comes the transition from the negative ‘logic of failure’, where policy is only admissible if it corrects inefficiencies in the static allocation of resources, towards the dynamic goal of enhancing the system’s ability to evolve. Second, the evolutionary-structuralist perspective (Dopfer et al., 2004; Lipsey et al., 2005) opens our field of vision to the problem of competitiveness not only for individual firms, but also at the level of industries, regions and countries. The competitiveness of aggregate entities thus becomes a meaningful and necessary concern of policy, generally characterized by the statistical moments of variables on economic performance in heterogeneous populations. Competitiveness policy is thus tightly embedded in a multi-layer system of enterprise, industrial and general framework policies. Third, the evolutionary framework invokes the political economy dimension of government action, such as the rules which determine the returns to productive vs rent-seeking activities, group con­ flicts, regulatory capture, or corruption. Their quality depends on the strength and integrity of the public institutions. To this, we must add the degree of development, which affects the kind of policies needed and the capacity to conduct them, as well as strategic concerns, such as the threat of escalating trade and subsidy wars. Finally, by taking the various advantages together, the dynamic approach better aligns the theoretical rationale for public interventions with the actual intentions and motivations of many of the actors responsible for particular policies in practice. The argument will proceed in four steps: Section 23.2 offers some basic considerations to clear the ground. Section 23.3 distinguishes the micro, meso and macro levels of develop­ ment for targeting public interventions. In Section 23.4 this is done for the different system functions that enable evolutionary change. The combination of both dimensions produces an integrated classification of competitiveness policies in Section 23.5. The resulting typology differentiates policies according to their characteristic functions, and target level and integrates them by their common purpose of enhancing the economic system’s capacity to evolve. Section 23.6 summarizes and concludes.

23.2

Basic considerations

Before turning to the core argument of this policy note, this section briefly addresses two potential misconceptions and resulting objections to the notion of competitiveness as defined earlier. First, in the context of growing ecological concerns about the limits to growth, the focus on rising real incomes requires an explanation that points to the very modern significance of Schumpeter’s emphasis on qualitative transformations. Second, to further clear the ground, Krugman’s influential critique of the notion of competitiveness for aggregate economies shall be refuted.

23.2.1

Qualitative change and growth ‘beyond GDP’

Evolving systems must be open to new opportunities2. Hence, to Schumpeter there are no definite boundaries to innovation, at least known, neither from running out of feasible 300

Evolutionary economic policy and competitiveness

technological changes, nor an ultimate saturation of demand. Like his teacher Friedrich Wieser, he believed in the common psychological trait of people continuously ‘discovering new directions of desire.’3 Because aspirations and demands tend to rise along with the attained standard of living, ‘satiety becomes a flying goal, particularly if we include leisure among consumers’ goods.’4 In Schumpeter’s time, the ecological limits of growth were not yet a general concern. However, their ever-growing importance today warrants a short reference to the Austrian theory of value. In his contribution to modern welfare economics, Friedrich Wieser stressed that ‘price is a social fact, but it does not denote the estimate put upon goods by society.’5 Wieser’s immediate concern was the difference between average utility as a measure of subjective well-being and marginal utility as a determinant of prices – for example, in the case of a monopoly raising prices to the disadvantage of consumers. Today, we may also think, for instance, of negative externalities from environmental degradation, or the accu­ rate measurement of welfare beyond the common national product, which is generally based on transaction prices.6 In case of conflict between subjective utilities and transaction values (i.e. prices), Wieser left no doubt as to the proper prioritization: ‘The highest principle of all economy is utility. Where value and utility come into conflict, utility must conquer; there is nothing in the nature of value which could give it the ascendency.’7 Schumpeter clearly agreed with his teacher’s view, for example, when stressing the importance of income dis­ tribution in the determination of social value.8 In face of the contemporary problems of growing inequality and degradation of the natural environment, one cannot doubt that the aspired transformation to more sustainable and ecological means of production would add to a society’s utility and well-being, and hence also to real income in terms of the quality of life people can afford. Consistent with our definition of competitiveness, such a transformation goes together with qualitative change and a rising standard of living, which points exactly at the deeper and very con­ temporary meaning of Schumpeter’s theory of economic development.9 Clearly, his vision reached beyond GDP, and so should our conceptions of competitiveness and competitive­ ness policy.

23.2.2

Krugman’s critique

According to Paul Krugman (1994, 1996), the concept of competitiveness applies only to individual firms. In contrast, nations or regions would not compete in any meaningful sense. But competition arises from scarcity, which, among other factors, can affect natural resources, capital, labor, human skills, or technological knowledge. Also, access to certain markets can be scarce, giving a natural advantage to firms in a location that is better integrated than others. The crucial question, therefore, is whether such scarcities only affect individual enterprises, households or workers. This can hardly be the case, since the relative abundance of the various factors of production, including knowledge, influences firms’ locational choices and their differential performance. In other words, relative scarcities at the aggregate level are the source of ‘comparative advantages’, which affect industrial location and specialization at the meso level. And when industries systematically vary in their productivity performance, differences in industrial specialization also affect a region’s overall per capita income. At a very fundamental level, the notion of competitiveness acknowledges that locations are not in the state of a unique, full employment perfect equilibrium (Fagerberg, 1996). 301

Michael Peneder

Instead, they regularly compete for activities with high value added as the source of high per capita incomes and hence material well-being. While zero-sum-games are indeed exceptions rather than the rule, the rivalry between different business locations can take various forms: First, locations compete directly for scarce resources, as is the case with the promotion of inward foreign direct investments. This carries a considerable potential of mutual conflict, e.g. when negotiating the terms of international agreements on investments and trade. Second, most often the competition between locations is indirect, trying to provide a favorable business environment in general, or fostering e.g. innovation and productivity growth with a focus on the particular needs of individual sectors. ‘Technology races’ are a case in point, where an economy’s better ability to earn rents from innovation also implies a greater command over scarce resources. Third, a peculiar kind of rivalry originates in the strive for political support and legitimation. By comparing their economic, social and en­ vironmental performance, numerous popular benchmarks exploit the governments’ ambi­ tion to score high in comparison to other countries or regions. By definition, the number of top ranks is limited, hence scarce and incites rivalry. To conclude, competitiveness at the aggregate level is a common and widely accepted concern of economic policy, grounded in various forms of actual competition, whether direct, indirect, or through a general com­ petition for political support and legitimacy

23.3

Ontology of change: Micro, meso and macro

Krugman’s critique of the concept of competitiveness fails to recognize the fundamental relationship between the micro, meso and macro levels of economic development that constitute the basic ontology of evolutionary change as a multi-level process.10 Aggregate populations thereby evolve through the changing composition of their individual entities which tend to associate and emerge into structure (Dopfer et al., 2004). From an evolu­ tionary perspective, the economy is generally not in perfect equilibrium with full employ­ ment. Instead, economies develop in a permanent succession of disequilibria. For such initial conditions, Figure 23.1 illustrates the connectedness between the micro, meso and macro levels of economic development with a schematic representation of how changes in competitiveness at one level systematically affect also the competitiveness at each of the other levels. For example, we may assume that all firms in a particular sector become more competitive either in terms of increased profitability, faster growth or improved odds of survival or entry of a new firm. Ceteris paribus and by mere aggregation, this will also raise the competitive performance at the meso level of the according industry in that location. Typical measures of performance would be the average profitability of the sector, the growth in sector output and the growth in the number of active firms or an increase in the sector’s revealed comparative advantage (RCA). By the same reasoning, the better per­ formance of that industry positively affects the aggregate performance at the level of individual countries or regions. The improved competitiveness may affect aggregate MFP growth, employment and hours worked as well as GDP per hour. Each of those changes will end up in an increase in GDP per capita and hence average income. But the causal effects do not only move from the micro to the meso and then the macro level through simple aggregation. For example, higher overall income feeds back to the individual firm via increased demand for final and intermediate goods, which will positively affect competitiveness at the firm level. In addition to this positive feedback from the macro to the micro level, the stronger industry performance also raises the potential for positive 302

Evolutionary economic policy and competitiveness

Growth

Structural change

+

+

Growth

Structural change

– Competition for specialised inputs

Demand

Industry Spillovers (Knowledge, specialised suppliers and labour, etc.)

– Competition for general inputs

Country and Region

Enterprise

Figure 23.1

Co-evolution of the micro, meso and macro levels of competitiveness.

spillovers (e.g. via knowledge diffusion, or specialized suppliers and labor), thus establishing also a positive feedback from the meso to the micro level. But there exists no perpetuum mobile, and even without assuming perfect equilibrium with full employment there will be negative feedbacks, or trade-offs, from the expanding use of scarce resources. For example, higher incomes at the macro level increase the demand and competition for general inputs of production such as labor, capital or natural resources, and thereby raise their prices. Similarly, the faster growth of a particular industry tends to increase the demand for and the prices of specialized inputs. Consequently, the tighter competition for common but scarce resources weakens the position of firms with a lesser ability to raise the necessary funds relative to those who can better afford to purchase inputs at the elevated price. These negative feedbacks from the macro and the meso level negatively affect the individual firm, but not necessarily the competitiveness of the industry as such. The reason is that higher input prices also foster structural change, driving the less com­ petitive firms out of the market and thereby increasing the share of those with a higher profitability, capacity to grow, etc. At least in the medium to long run, this change in the composition of firms may improve the performance of the industry at the meso level. Relatedly, the competition for general inputs may foster structural change at the meso level and shift the composition of production towards the more productive sectors that create more value per inputs and hence can afford to pay higher prices. In a nutshell, the micro, meso and macro levels of economic development are inextricably interwoven, i.e. inter­ dependent parts of the same reality. The evolutionary-structuralist agenda has produced a wealth of literature that provides further specificity to this very general relationship between the micro, meso and macro levels 303

Michael Peneder

of development. Starting from the bottom upwards, at the micro level the individual pro­ cesses of creating, maintaining and capturing value from competitive advantage are highly situational and idiosyncratic. While novelty originates in the opportunity-seeking creativity of individual and corporate entrepreneurship, organization and management define the structural context of incentives and decision making, determining which ideas get autho­ rized and funded. The dynamic capabilities approach has particularly embraced the evolu­ tionary perspective at the level of strategic management. Merging the idea of Schumpeterian competition with resource based theories of the firm in the tradition of Penrose (1959), the focus is not on efficient allocation or contracts, but on how firms create and capture value within fast moving environments (Teece et al., 1997; Pitelis and Teece, 2009). One core finding is that distinctive competences cannot, in general, be acquired through market transactions, but the firm itself must search, build, integrate and continuously reconfigure them. Within complex environments this involves substantial sunk costs from long-term commitments to particular competence domains with the consequence of organizational inertia. Since search tends to be local, typically in the neighborhood of existing knowl­ edge, competencies and routines, history and organization-specific processes of capability accumulation shape the evolution of firms and entire industries (Helfat and Winter, 2011; Helfat, 2018). At the meso level of firm populations, the study of corporate demography and organi­ zational ecology pursues a deliberate evolutionary approach (Hannan and Freeman, 1989; Carroll and Hannan, 2000, Carroll and Khessina (2019). One of its most interesting hypotheses concerns resource partitioning among large and small firms, which is more effective in discovering and exploiting varied niches from segmented markets and heter­ ogenous consumer tastes. It is a compelling instance of how the industry-level composition of heterogenous firms impacts on the aggregate creation of value. While the micro level explains the sources of variation, consumer tastes and preferences define the selection en­ vironment. Of course, these are not independent, but also conditioned by their perception of what variety and quality of goods the firms can offer. In short, the micro and meso level coevolve with qualitative transformations that relate to changes in the composition of lowerlevel structural characteristics. If we finally move towards linking the meso and macro levels of economic activity, Pasinetti (1981, 1993) provided the canonical model of structural economic dynamics. Structural variations enter via sectoral differences in technological change and demand elasticities. Both sources are exogenous, but the model elaborates their interaction within a closed economy characterized by macro-economic resource constraints. The main mecha­ nism shows that productivity growth in a particular sector leads to a decline of its relative prices, whereas consumers spend the according gains of real per capita income also on other sectors, depending on their respective income elasticities of demand. While technological change drives the growth of real income, different demand elasticities determine the sectoral composition of production, which in turn affects the weight of further productivity gains of a sector in aggregate growth.11 The upshot is that aggregate populations must evolve through structural changes in favor of more productive activities, and thereby raise the average ability to alter the given material constraints. Consistent with our definition of competitiveness, this directly relates to an economy’s capacity to earn high and sustainable per capita incomes by adapting to qualitative transformations of the system as well as actively managing them. 304

Evolutionary economic policy and competitiveness

23.4

System functions: The logic of public intervention 23.4.1 Rationalities of failure

Theoretic rationales of economic policy start from the basic dichotomy of arguments for or against public intervention. On one side of the debate we find arguments in support of free markets and laissez-faire, who lay much emphasis on ‘government failures’, typically cast in terms of agency problems that make room for bureaucratic inefficiencies or regulatory capture by vested interests.12 On the other side of the debate, the supporters of public interventions invoke either the common rationales of ‘market failure’, or alternatively refer to notions such as ‘system failure’ and ‘strategic failures’ in policy making.13 These ratio­ nalities of failure are generally well defined and obviously have their true points: Government failure is omnipresent. If taken literally, market failure is equally ubiquitous, simply because the idea of perfect competition represents a hypothetical ideal state that is generally untenable in real business. Also, the notion of system failure correctly addresses blind spots in the previous arguments, for instance, by paying attention to barriers of cooperation and knowledge flows between relevant actors and organizations. Finally, strategic failure can result from the government’s lack of understanding that there are multiple potential equilibrium solutions or how their policy choices can affect their actual realization. However, because these rationales are so ubiquitous, they regularly apply simultaneously. Despite their theoretical rigor, in practice, they regularly leave policy­ makers to make informed judgemental decisions. Moreover, it is particularly revealing that all the rationales refer to failures in order to legitimize, motivate, or discard public interventions. Can we think of other areas in which we accept such a logic of failure to motivate human actions? Probably not. It is rather a very peculiar attitude of our profession that accordingly calls for an explanation. In short, economists learn to accept the ideal of hypothetical perfect states as a normative bench­ mark. This directly relates to our preoccupation with welfare economics. What has been developed there for good analytical purposes, has by means of routinized exposure in the textbooks and habit in the scientific discourse become so deeply engrained in our canon, that it is mostly taken for granted without initiating much further reflection. Or who, in the realm of policy debate, has taken seriously the implication of Arrow’s (1950) impossibility theorem, which demonstrated that neither political voting nor the market mechanism can create optimal social choices in the sense of a rational and consistent aggregation of the preferences of sovereign individuals: ‘The failure of purely individualistic assumptions to lead to a well-defined social welfare function means, in effect, that there must be a divergence between social and private benefits if we are to be able to discuss a social optimum. Part of each indvidual’s value system must be a scheme of socio-ethical norms, the realization of which cannot, by their nature, be achieved through atomistic market behavior’ (Arrow, 1950: 343). Other paradoxes and inconsistencies in the aggregation of individual preferences have since cropped up.14 For instance, Jackson and Yariv (2014, 2015) point at the systematic bias towards the present, and showed that with heterogeneous time preferences any Pareto and non-dictatorial method of aggregation must either be time-inconsistent or intransitive. To conclude, the frequent assumption of an all-powerful, omniscient and benevolent dictator, who appears in numerous policy models, is not that innocuous as generally perceived. It does not, as many belief, merely abstract from the obvious difficulties of 305

Michael Peneder

policy implementation, but avoids to admit the theoretical impossibility to determine the public intervention by the benchmark of a reasonably defined and unique social welfare optimum. This is even more the case when we turn to dynamic and open sys­ tems. Their normative yardsticks of hypothetical perfect states are ill-defined, and thus the heuristic of ‘failure’ is a poor basis for public intervention. In contrast, competi­ tiveness policy requires a dynamic rationale instead of the traditional static argument of allocative efficiency. Based on the distinct heuristic of economic development, the key question to start with is not what ‘failure’ a policy needs to fix, but what goals it wants to achieve.

23.4.2

Ability to evolve

As defined in the beginning, competitiveness policy aims to foster economic development, which is tantamount to the objective of enabling and molding evolutionary change. At the most general level, and consistent with a wide array of authors,15 evolutionary change is characterized by the simultaneous interplay of the three elementary principles of var­ iation, cumulation and selection. They are not meant as an analogy from the natural sciences, but represent a higher level of abstraction (a meta-theory), which can charac­ terize the time behavior of many different systems and forms. It, therefore, goes without saying that the evolution of a socio-economic system is not a process of natural selection, but one of cultural evolution (Dopfer, 2016; Dopfer and Nelson, 2018). But irrespective of the manifold differences among their specific realizations, no system can evolve if either of the three functions is missing. In short, novelty and the according variety are requisite to any change, accumulation introduces the dimension of time and renders the system dynamic,16 and finally, selection channels the process towards altering constraints. The selection environment is where scarcity, and hence competition and competitiveness, come into play. Table 23.1 further illustrates the requisite nature of these functions for evolutionary change. None of them raises much analytic interest, if considered alone. A common ex­ ample of pure variation is white noise, where any observation yt = t , and t is an inde­ pendently distributed random variable of zero mean. Without selection, there is no scarcity and hence no economic interpretation. The same applies to mere accumulation, for instance, the deterministic but blind growth by a constant factor a, where y (t ) = y0 e at . Table 23.1 A General Characterization of System Dynamics Variety → Change Stasis Blind growth White noise Static equilibrium Random walk/drift Steady state growth Evolutionary change

Stochastic − − + + Structural + − +

Source: Peneder (2001).

306

Cumulation → Time

Selection → Direction

− + − −

+ − − +

+ + +

− + +

Evolutionary economic policy and competitiveness

Finally, pure selection without variety has nothing to operate upon. We may call this state stasis. For illustration, consider the general selection dynamics dxi = xi (fi dt

),

(23.1)

where xi is the frequency of type i , which increases (decreases), if its fitness fi is higher (lower) than the average fitness of a population . Without variety, fi must be equal to . Hence, there won’t be any change in the frequencies, or other dynamic behavior. Still, this description can be of practical importance, since systems may get locked into such a situation, if selection has previously consumed all of its requisite variety and no (endogenous) source of novelty refuels the process. When two of the elementary principles apply, complexity increases and some familiar characterizations emerge. First, we consider variation in combination with accumulation, while no principle of selection applies. Again, the lack of selective constraints implies that this system is not subject to any economic interpretation. The typical example would be a random walk. In the simplest case, the time series is determined by the expression yt = yt 1 + t with t being a random source of variation and yt 1 representing the cumulative and time-dependent nature of the system. If we add a constant trend component d, the process yt = yt 1 + d + t is called random drift. We can similarly imagine a system in which only the two elements of variation and selection interact, but no accumulation takes place. This would correspond to a stationary state such as in the familiar static equilibrium models. The lack of accumulation means that the system is invariant with respect to time. Variations only cause fluctuations around a certain equilibrium configuration determined by the selective constraints. The crucial assumption is that at any point in time, the probability distribution p of the variable yt remains the same: p (yt ) = p (yt + m ). Perfect competition achieves exactly that by assuming the strongest possible mechanism of selection, which instantaneously eliminates any deviations from the optimal equilibrium solution. Steady state growth is a convenient example of the interaction of cumulation with selection. It is the foremost starting point for breaking through from a stationary state to a dynamic system and implies that in equilibrium the various quantities grow at constant rates. Stochastic variations may exist (as with the aforementioned stationary state), but consistent with their macroeconomic focus, the models do not depict variation with a continuing effect in the sense of structural changes.17 Theories of endogenous growth also rely on the framework of steady-state equilibrium analysis, thereby eschewing the even more complex dynamics of evolutionary change.18 This brings us to our final characterization of evolutionary change as the simultaneous interplay of all three functions. At its most general, consider a constant population of x = (x1, … xn ) units of selection, let’s say individual firms. These are carriers of either of i = 1,...,N distinct information sets, let’s say different production systems. Firms may change from system j to system i with probability qji . The transition matrix Q = [qij ] is stochastic and quadratic (n × n). Innovation, i.e. the introduction of a novel production system, corresponds to a change of frequency xi from 0 to 1. Other changes in the fre­ quencies represent the further diffusion, decline, or extinction of a production system. Selection occurs over the simplex Sn , i.e. in=1 xi =1, and production systems have different fitness values fi . Changes in the frequency of particular production systems depend on their 307

Michael Peneder

advantage or disadvantage relative to the average fitness of the population = in=1 xi fi . If fitness values depend on the frequency distribution of production systems x , a fundamental equation of evolutionary change (Nowak, 2006) can be expressed as follows:

dxi = dt

n j=0

xj f j ( x ) qji

xi .

(23.2)

Such models typically offer no closed solution, but require the use of analytic simulations (e.g. Caiani et al., 2014; Dosi et al., 2014). Note that the neoclassical equilibrium with perfect competition characterizes a special case, where all firms simultaneously know about a new technology, can adopt them (or enter and exit the market) without cost, and must immediately do so. Otherwise, consumers can instantaneously and without cost shift to the most competitive rival. Since these assumptions don’t allow for any meaningful differences in the relative fitness of the firms, the entire dynamic of this equation, i.e. the structural changes in the frequency distribution of the population won’t occur. In contrast, with evolutionary change firms must continuously search their fitness land­ scape, which is a time-consuming process of learning, often by means of costly trial and error. Hayek (1945) already characterized market competition as a discovery process, which effec­ tively co-ordinates the largely decentralized knowledge about supply and demand. But thinking of the above selection equation, market competition additionally fosters learning about one’s own competitive advantage, helping to specialize in activities that congrue with one’s actual relative strengths and weaknesses. In their pursuit of favorable resource niches, populations thus tend to move uphill’, i.e. find or adopt information sets with higher fitness values. Since the variety of behavior is not instantaneously selected away by any rule of perfection, novel ideas enjoy a certain margin of error. This permits experimentation and the accumulation of more complex information sets through learning. Selection still operates in favor of production systems that are more effective in altering given scarcities. In the sphere of cultural evolution, this means foremost the deliberate search and adoption of better rules and practices through learning. Where such capabilities are constrained, differential growth, or the dynamics of entry and exit will take its place (Nelson and Winter, 1982; Metcalfe, 1998).

23.5

Fitting the pieces: An integrated classification

Combining the target levels and system functions of economic development, one can organize a fairly comprehensive variety of different public interventions into a concise and meaningful typology of competitiveness policies, as shown in Table 23.2. To begin with the distinction between the micro, meso and macro levels of development, (i) enterprise policies address individual firms, whereas (ii) structural policies target intermediate levels of aggregation, such as specific industries, technologies, clusters, or networks and (iii) frame­ work policies comprise economy-wide regulations and institutions, infrastructure, as well as public interventions for macroeconomic stabilization. At each target level, concrete policies aim to serve the basic system functions of (i) resurrecting requisite varieties by means of novelty, (ii) the accumulation of productive resources, or (iii) shaping the selection en­ vironment through markets and regulations. The various elements are interdependent and co-evolve within complex, path-dependent and non-deterministic processes (Arthur, 2014). Though the different functions and target levels must consequently overlap in terms of

308

Evolutionary economic policy and competitiveness Table 23.2 General Typology and Examples of an Integrated Competitiveness Policy Target level

System functions Novelty

Micro

Innovation and start-up policy

Meso

Technology policy

Macro

Research policy

Resources Enterprise policies Subsidies, micro-credit, venture capital, etc. Industrial policies Targeted investment and diffusion schemes, regional clusters Framework policies Monetary and fiscal policy, infrastructure, education

Markets and regulation SWFs, public procurement Competition policy, trade policy, sector regulations Economic integration, environmental, social, and labor regulations

Source: Peneder (2017).

specific organizational bodies and institutional arrangements, this is not to detract from their characteristically distinct logic of public intervention. A few examples may illustrate the point. If we begin with the system function of introducing novelty to the system, one can apply the common convention of distinguishing between research, technology and innovation policies in terms of our micro, meso and macro structure of the target levels. Research policy then addresses the macro-level framework of R&D without directly discriminating between particular firms, sectors or technologies. The emphasis is typically more on basic research and scientific excellence rather than business applications and immediate economic returns. Political savvy in terms of the ability to promote an agenda and to assert one’s claim for public resources, legal expertise together with a thorough understanding of how rules and regulations affect the incentives within the research community are distinctive competences needed at this level of policy making. In contrast, technology policies target particular fields of activity, such as certain general purpose technologies, and directly intervene in their favor. Typically, they address basic and applied research. Pursuing the same overall function of introducing novelty to the system, the target communities, instruments, tools and required expertise are nevertheless fundamentally different from the former activities at the macro level. Strategic planning and the ability to set priorities among different technology fields are distinctive competences. For the selection of proper targets, policy puts much emphasis on scienceindustry relationships and the involvement of stakeholders. Finally, at the micro level, innovation and start-up policies address individual enterprises, which may, for example, apply for grants, preferred loans, guarantees, or equity-related instruments provided by specialized promotion agencies. The focus accordingly shifts from basic to applied research and the establishment of new enterprises. A key competence is in handling individual projects, e.g. by offering a fair, efficient and accurate selection among submissions. This requires specific process knowledge and a reasonable ability to understand and evaluate heterogeneous projects. If we turn to the second general system function, i.e. the accumulation of productive resources, investment in its various forms is the key concern and framework policies at the macro-level dominate the picture. Monetary policy is relevant given that the non-neutrality

309

Michael Peneder

of money with respect to real output directly relates to the evolutionary-structuralist per­ spective.19 Fiscal policy uses taxes and public spending to affect, for example, returns on investment, disposable income, or incentives to work, as well as to stabilize expectations about future demand. At the meso level, strategic considerations can trigger public spending targeted to specific locations and activities, such as the provision of advanced transporta­ tion systems, communications networks or specialized educational facilities as well as targeted programs to foster the diffusion of new technologies, each adapted to the needs of local industries and clusters of related activities. At the micro level, most policies that target indi­ vidual enterprises relate to the funding of investments, for instance by means of subsidized loans, guarantees or equity instruments. Many initiatives target small and medium-sized enterprises, venture capital, exports, or underprivileged groups (e.g. minorities, women, people in distressed areas). In a dynamic perspective, the tricky challenge for policy is to help kick-start a process, but to get out of the way when private initiatives start to develop. Finally, public policy shapes the selection environment in many ways. At the macro level, far-reaching choices regard the kind and degree of economic integration. While not easily explained in terms of market failures, levelling the selection environment among otherwise segmented markets is arguably one of its distinctive goals.20 In addition, there are all kinds of social, labor, environmental and other regulations that shape the selection environment by defining by what means firms are allowed or not allowed to compete. Many policies affect the selection environment at the meso-level. For instance, trade agreements specify very detailed rules for different industries. The same applies to the many detailed product regulations. Moreover, competition policy deliberately aims to enhance the efficiency of selection in the economic system and seeks to protect consumers and potential new entrants from the abuse of market power by incumbent firms. One of its biggest chal­ lenges is to balance the trade-off with the function of introducing novelty, where temporary monopoly rents from innovation are the primary incentive to invest resources in R&D and related activities.21 Policies that interfere with the selection process by addressing individual enterprises are a particularly discriminating form of intervention. State ownership and picking-the-winners type industrial policies are notorious cases that recall many historical examples of government failure. But in the wake of globalization, the rapid growth and ex­ pansion of sovereign wealth funds has also altered the picture in recent years. Public pro­ curement is another notable instance. It draws least attention, where it is most common, i.e. with regard to favoring local content in the regular procurement of goods and services by (local) governments and public organizations. From a dynamic perspective, maintaining a varied and differentiated ecology of firms can be a valid goal of regional development, and especially in distressed regions may well dominate pure efficiency considerations. However, the economic cost of restricting one’s supply base by way of privileges to firms with a local representation increases with economies of scale and technological complexity.

23.6 Summary and conclusions This brief note has presented an evolutionary concept of competitiveness policy as driver of Schumpeterian development, characterized by the combination of growing real incomes and qualitative changes of the socio-economic system. Proposing a distinct dynamic logic of public intervention, it aims to reconcile the theoretic rationales of economic policies with the actual concern of most public agencies in practice. In summary, we can highlight the fol­ lowing findings: 310

Evolutionary economic policy and competitiveness

• First, the analysis has shown that the conventional critique on the notion of competi­ tiveness for aggregate economies ignores the fundamental relatedness between the micro, meso and macro levels of development. Rather than passively adapting to given factor endowments and the according comparative advantages, locations compete for favorable business conditions to generate higher incomes and living standards by creating and capturing more value within the global systems of production. • Second, economists show a peculiar attachment to ‘rationalities of failure’, be it either of markets, governments, or systems. It originates in our habit to accept hypothetical perfect states as normative benchmarks, inherited from the canon of static welfare optimization. This stands in contrast to a dynamic logic of intervention, which should target the functions that an open system must accomplish. • Third, by matching different target levels with the functional principles of evolutionary change, the resulting classification of economic policies allows to better coordinate and direct them toward their common development goals. More specifically, for each type of intervention, it substitutes the conventional ‘rationality of failure’ in terms of allocative efficiency by its particular dynamic function to enhance the system’s capacity to evolve. While many economic policy measures can also be reasoned with the traditional rationale of market failure, the situation resembles a tilted image: Once one acknowledges the dynamic function of these policies and their orientation towards economic development, the efficiency-based logic of failure appears uncomfortably forced. Moreover, it cannot provide an integrated perspective. Allocative efficiency is generally not their constitutive purpose, but more an intellectual bracket to align common policy sense with the theoretical canon. The uneasiness of this match has greatly contributed to the widening gap between politics and economic research. In contrast, the contribution of the various policies to a wellperforming socio-economic system, reasonably efficient in the short run, but more impor­ tantly capable of development in the long run, can better provide for a unifying goal. It is also more consistent with how people working in the various policy agencies perceive themselves and their contribution to society. The switch of perspective from the canonical rationalities of failure towards the system’s ability to evolve thus creates the opportunity for a novel, more realistic and better-integrated understanding of economic policy. But what does all this say about when governments should or should not actually inter­ vene? Aren’t we losing the clear guidelines typically offered by the simpler rationale of market failure? In short, the dynamic perspective is necessarily more comprehensive. It offers more complex answers but in reply to deeper questions. However, the core of the market failure argument is easily taken on board by the even more straightforward rule of opportunity costs: If private markets are more efficient than governments in accomplishing a certain task, then don’t waste public resources on it that can create more value in other uses. Given the scarcity of public resources, a positive net gain in welfare thus need not be sufficient to justify a particular initiative. Instead, governments must set priorities according to their anticipation of the relative benefits and costs with respect to their overall development goals. Finally, an integrated and dynamic perspective cannot ignore the fact that both synergies and conflicts arise between different dimensions of competitiveness. To ensure that longterm goals are not neglected in favor of the ever more pressing short-term needs, compet­ itiveness requires the simultaneous consideration and balancing of different time horizons, as shown in Table 23.3.22 In the short term, the focus is on the system’s capacity to adapt to changing framework conditions. Imbalances should be avoided and macroeconomic 311

Michael Peneder Table 23.3 Different Time Horizons of Competitiveness Policy: Selected Examples Time horizon

Change in framework conditions

Adaptive system response

Policy objectives

Short-term

Fluctuations of demand, Profit margins, wages, Attenuation of cyclical exchange rates, prices, public spending, fluctuations and crises etc. monetary policy, etc. (jobs, production, prices) Medium-term Technology (standards etc.) Innovation, education Productivity, full empGlobalization (e.g. value and training, investment, loyment, market shares, chains) internationalization resource efficiency Long-term Artificial intelligence, Education, public infraHigh real incomes, social societal demands, structure, environmental inclusion and participation, climate change and health decarbonization standards, etc. Source: WIFO.

stability maintained by either constraining or mobilizing current spending. Typical em­ pirical measures are, for example, real effective exchange rates, unit labor costs, inflation, or the current account balance. Monetary policy, fiscal policy and wage policy are among the most important macroeconomic tools of public intervention. In the medium term, the core objectives regard the dynamics of the economic system, which is reflected both in pro­ ductivity growth and in the goals of full employment, high market shares in exports and improvements in energy and resource efficiency. Key determinants include innovation, investments, internationalization, as well as competition and regulation. Finally, in the long term, the quality of life must be at the center of attention. Priorities are sustainable high real incomes, social inclusion and participation as well as the improvement of the natural en­ vironment and the avoidance of irreversible climate change. Advancing the implementation of these goals simultaneously with all other objectives thus remains one of the greatest challenges of competitiveness policy.

Notes 1 This contribution is an updated and condensed discussion of a topic that the author has pre­ viously addressed in more detail and from different perspectives in Peneder (2001, 2017) or Peneder and Rammer (2018). It owes greatly to many inspiring debates at various occasions, most notably Kurt Dopfer’s Vienna Seminar on Evolutionary Economics and the Thematic Platform on Competitiveness of the Austrian Institute of Economic Research (WIFO). The usual disclaimer applies. 2 Schumpeter (1942/1950, p. 118). 3 Wieser (1914, p. 26). 4 Schumpeter (1942/1950, p. 131). On logical grounds, one cannot preclude satiation, and the recurrent fear of secular stagnation is a reasoned possibility that Schumpeter acknowledged. But he also argued that the historical record of technological change suggests otherwise ( Schumpeter, 1939, p. 1035; 1942/1950, pp. 111ff). 5 Wieser (1889, p. 64f). 6 Jones and Klenow (2016), Stiglitz et al (2018). 7 Wieser (1889, p. 54). 8 “Not only must the sum of individual wealth be given, but also its distribution among individuals. Marginal utilities do not depend on what society as such has, but on what individual members have. [] The distribution of wealth is important for determining values and shaping production,

312

Evolutionary economic policy and competitiveness

9 10 11

12 13 14 15 16 17

18 19 20

21 22

and it can even be maintained that a country with one and the same amount of general wealth may be rich or poor according to the manner in which that wealth is distributed” ( Schumpeter, 1909, p. 214f). Peneder and Resch (2021). In philosophy, an ontology is about the basic structures of reality and their classification. Relatedly, in information science, it defines a set of concepts and categories that represent a subject. In the meantime, manyfold extensions of Pasinetti’s multi-sector model have flourished. To give an example, Araujo and Lima (2007) highlight how the combination of sectoral specialization and aggregate income affect the elasticity ratio of export vs import demand and thereby the economy’s overall growth path. Among further examples, see Cimoli and Porcile (2011) or Araujo and Trigg (2015). Tullock (1967), Krueger (1974). See, e.g. Smith (2000) or Cowling and Tomlinson (2000). Blackorby and Donaldson (1990). See Veblen (1898), Nelson and Winter (1982), Hodgson (1993, 2002), Metcalfe (1994, 1998), Aldrich et al. (2008), or Winter (2014). For a critical view, see Witt (2008), Buenstorf (2006) and Cordes (2006). The emphasis on ‘accumulation’ is owed to the interest in economic development and growth of real income. Other disciplines use different terms, such as ‘reproduction’ in biology, ‘integration’ in sociology ( Luhmann, 1997), or ‘retention’ in institutional economics ( Hodgson, 1993). Note that Hicks refused to consider them dynamic: “I do want to say that perfect foresight models, such as steady state models, really are static. Although there are differences between one moment of time and another, they have so much in common that the thing really remains static” (quoted in Klamer, 1989, p. 173). See Aghion and Howitt (2009) for models of innovation as an endogenous driver of Schumpeterian growth. This is nevertheless different from Schumpeterian development, with its additional em­ phasis on structural change among heterogeneous populations. Peneder and Resch (2021). Integration strongly interacts with the other functions. Besides the productivity enhancing effects of increased competition, the larger markets offer bigger opportunities for economies of scale and spe­ cialization, hence raising the incentives for investment, including innovation. Policies range from pegging one’s currency to that of a major trading partner, various multi- and bilateral agreements on trade, FDI, or intellectual property rights, up to the EU’s Single Market Programme. Aghion et al. (2005), Peneder and Woerter (2014), or Pyka and Nelson (2018). The present distinction by time horizons originates in the joint work with Thomas Url, Angela Koeppl, Peter Mayerhofer and Thomas Leoni; see https://www.wifo.ac.atjartprj3wifomain.jart? rel=en&content-id=1568136628866&reserve-mode=active.

References Aldrich, H.E., Hodgson, G.M., Hull, D.L., Knudsen, T., Mokyr, J., Vanberg, V. 2008. In Defence of Generalized Darwinism, Journal of Evolutionary Economics 18, 577–596. Aghion, P., Bloom, N., Blundell, R., Griffith, R., Howitt, P. 2005. Competition and Innovation: An Inverted-U Relationship, Quarterly Journal of Economics 120, 701–728. Aghion, P., Howitt, P. 2009. The Economics of Growth, MIT Press, Cambridge MA. Araujo, R.A., Lima, G.T. 2007. A Structural Economic Dynamics Approach to Balance-of-PaymentsConstrained Growth, Cambridge Journal of Economics 31, 755–774. Araujo, R.A., Trigg, A. 2015. A Neo-Kaldorian Approach to Structural Economic Dynamics, Structural Change and Economic Dynamics. Arrow, K., 1950. A Difficulty in the Concept of Social Welfare, Journal of Political Economy 58 (4), 328–346. Arthur, W.B. 2014. Complexity and the Economy, Oxford University Press, Oxford. Blackorby, C., Donaldson, D. (1990), The Case Against the Use of the Sum of Compensating Variations in Cost-Benefit Analysis, Canadian Journal of Economics 23 (3), 471–494.

313

Michael Peneder Buenstorf, G. 2006. How Useful Is Generalized Darwinism as a Framework to Study Competition and Industrial Evolution?, Journal of Evolutionary Economics 16 (5), 511–527. Caiani, A., Godin, A., Lucarelli, S. (2014), A Stock Flow Consistent Analysis of a Schumpeterian Innovation Economy, Metroeconomica 65 (3), 397–429. Cantner, U., Pyka, A. 2001. Classifying Technology Policy from an Evolutionary Perspective, Research Policy 30 (5), 759–775. Carroll, G.R., Hannan, M.T. 2000. The Demography of Corporations and Industries, Princeton University Press, Princeton. Carroll, G.R., Khessina O.M., 2019. Organizational and Corporate Demography, in: Poston, D.I.Jr. (ed.), Handbook of Population (Second Edition), 521–554, Springer Nature, Cham, Switzerland. Cimoli, M., Porcile, G. 2011. Global Growth and International Cooperation: A Structuralist Perspective, Cambridge Journal of Economics 35, 383–400. Cordes, C. 2006. Darwinism in Economics: From Analogy to Continuity, Journal of Evolutionary Economics 16 (5), 529–541. Cowling, K., Tomlinson, P.R. 2000. The Japanese Crisis -- A Case of Strategic Failure?, Economic Journal 110, F358–F381. Dopfer, K. 2016. Evolutionary Economics, in: Faccarello, G., Kurz, H.D. (eds.), Handbook of the History of Economic Analysis, Vol. III Development in Major Fields of Economics, Chapter 14, Edward Elgar, Cheltenham UK. Dopfer, K., Foster, J., Potts, J. 2004. Micro--Meso--Macro, Journal of Evolutionary Economics 14 (2), 263–279. Dopfer, K., Nelson, R.R., 2018. The Evolution of Evolutionary Economics, in: Nelson, R.R., Dosi, G., Helfat, C.E., Pyka, A., Winter, S.G., Saviotti, P.P., Lee, K., Malerba, F., Dopfer, K. (eds.), Modern Evolutionary Economics, 208–229, Cambridge University Press, Cambridge UK,. Dosi, G., Napoletano, M., Roventini, A., Treibich, T. 2014. Micro and Macro Policies in the Keynes + Schumpeter Evolutionary Models, OFCE Working Paper 2014-19. Fagerberg, J. 1996. Technology and Competitiveness, Oxford Review of Economic Policy 12 (3), 39–51. Hannan, M.T., Freeman, J. 1989. Organizational Ecology, Harvard University Press, Cambridge MA. Hayek, F.A. 1945. The Use of Knowledge in Society, The American Economic Review 35, 519–530. Helfat C.E. 2018. The Behavior and Capabilities of Firms, in: Nelson R.R., Dosi G., Helfat C.E., Pyka A., Winter S.G., Saviotti P.P. Lee K., Malerba F., Dopfer K. (eds.), Modern Evolutionary Economics, 85–103, Cambridge University Press, Cambridge UK. Helfat C.E., Winter S.G. 2011. Untangling Dynamic and Operational Capabilities: Strategy for the (N)ever-changing World, Strategic Management Journal 32, 1243–1250. Hodgson, G.M. 1993. Economics and Evolution. Bringing Life Back to Economics, Polity Press, Cambridge UK. Hodgson, G.M. 2002. Darwinism in Economics: From Analogy to Ontology, Journal of Evolutionary Economics 12, 259–281. Irwin, D.A. 2004.The Aftermath of Hamilton’s “Report on Manufacturers”, Journal of Economic History 64 (3), 800–821. Jackson, M.O., Yariv, L. 2014. Present Bias and Collective Dynamic Choice in the Lab, American Economic Review 104 (12), 4184–4204. Jackson, M.O., Yariv, L. 2015. Collective Dynamic Choice: The Necessity of Time Inconsistency, American Economic Journal: Microeconomics 7 (4), 150–178. Jones, C.I., Klenow, P.J. 2016. Beyond GDP? Welfare across Countries and Time, American Economic Review 106 (9), 2426–2457. Klamer, A. 1989. An Accountant Among Economists: Conversation with Sir John R. Hicks, Journal of Economic Perspectives 3 (4), 167–180. Krueger A.O. 1974. The Political Economy of the Rent-Seeking Society, American Economic Review, 64 (3), 291–303. Krugman, P. 1994. Competitiveness: A dangerous obsession, Foreign Affairs 73 (2), 28–44. Krugman, P. 1996. Making Sense of the Competitiveness Debate, Oxford Review of Economic Policy 12 (3), 17–25. Lipsey, R.G., Carlaw, K.I., Bekar, C.T. 2005. Economic Transformations, Oxford University Press, Oxford. Luhmann, N. 1997. Die Gesellschaft der Gesellschaft, Suhrkamp, Frankfurt am Main.

314

Evolutionary economic policy and competitiveness Metcalfe, S.J. 1994. Evolutionary Economics and Technology Policy, The Economic Journal 104 (425), 931–944. Metcalfe, S.J. 1998. Evolutionary Economics and Creative Destruction. The Graz Schumpeter Lectures, Routledge, London. Nelson, R.R. 2009. Building Effective Innovation Systems’ versus Dealing With Market Failures’ as Ways of Thinking About Technology Policy, in: Foray D. (ed.), The New Economics of Technology Policy, 7–16, Edward Elgar, Cheltenham UK. Nelson, R.R., Winter, S.G. 1982. An Evolutionary Theory of Economic Change, Belknap Press, Cambridge MA. Nowak, M.A. 2006. Evolutionary Dynamics: Exploring the Equations of Life, Belknap, Cambridge MA. Pasinetti, L.L. 1981. Structural Change and Economic Growth: A Theoretical Essay on the Dynamics of the Wealth of Nations, Cambridge University Press, Cambridge UK. Pasinetti, L.L. 1993. Structural Economic Dynamics, Cambridge University Press, Cambridge UK. Peneder, M. 2001. Entrepreneurial Competition and Industrial Location, Edward Elgar, Cheltenham UK. Peneder, M., 2017. Competitiveness and Industrial Policy. From Rationalities of Failure Towards the Ability to Evolve, Cambridge Journal of Economics 41, 829–858. Peneder, M., Rammer, C., 2018. Measuring Competitiveness, Report to the European Commission, DG GROW, 2018, WIFO, Vienna. Peneder, M., Resch 2021. Schumpeter’s Venture Money, Oxford University Press, Oxford. Peneder, M., Woerter, M. 2014. Competition, R&D and Innovation: Testing the Inverted-U in a Simultaneous System, Journal of Evolutionary Economics 24 (3), 653–687. Penrose, E.T. 1959. The Theory of the Growth of the Firm. John Wiley & Sons, New York. Pitelis, C.N., Teece, D.J. 2009. The (New) Nature and Essence of the Firm, European Management Review 6, 5–15. Pyka, A., Nelson, R.R. 2018. Schumpeterian Competition and Industrial Dynamics, in: Nelson, R.R., Dosi, G., Helfat, C.E., Pyka, A., Winter, S.G., Saviotti, P.P. Lee, K., Malerba, F., Dopfer, K. (eds.), Modern Evolutionary Economics, 104–128, Cambridge University Press, Cambridge UK. Schumpeter J.A. 1909. On the Concept of Social Value, Quarterly Journal of Economics (Feb.), 213–232. Schumpeter, J.A. 1911. Theorie der wirtschaftlichen Entwicklung, 4th edition. Duncker & Humblot, Berlin. Schumpeter, J.A. 1939. Business Cycles, A Theoretical, Historical, and Statistical Analysis of the Capitalist Process, McGraw Hill, New York. Schumpeter, J.A. 1942/1950. Capitalism, Socialism and Democracy, 3rd edition, Harper & Row, New York. Smith, K. 2000. Innovation as a Systemic Phenomenon: Rethinking the Role of Policy, Enterprise & Innovation Management Studies 1 (1), 73–102. Stiglitz J.E., Fitoussi J-P., Durand M. 2018. Beyond GDP. Measuring What Counts for Economic and Social Performance, OECD, Paris. Teece, D.J., Pisano, G., Shuen, A. 1997. Dynamic Capabilities and Strategic Management, Strategic Management Journal 18 (7), 509–533. Tullock, G. 1967. The Welfare Costs of Tariffs, Monopolies, and Theft, Western Economic Journal 5 (3), 224–232. Wieser, F. 1889/1893. Der natürliche Werth, Vienna (English translation as ‘Natural Value’, London: Macmillan). Wieser F., 1904. Der Geldwert und seine geschichtlichen Veränderungen, Zeitschrift für Volkswirtschaft, Sozialpolitik und Verwaltung, XIII, Wien. Winter, S.G. 2014. The Future of Evolutionary Economics: Can We Break Out of the Beachhead? Journal of Institutional Economics 10 (4), 613–644. Witt, U. 2008. What Is Specific About Evolutionary Economics? Journal of Evolutionary Economics 18, 547–575.

315

24 SMART SPECIALISATION Dominique Foray

24.1

Introduction

The aim of this chapter is to convey a coherent vision of the policy approach evoked by the term “smart specialisation strategy” (S3) and explore and elaborate the requirements and implications that are consistent with giving operational content to this concept. Although, certainly, there are other conceptual framework and corresponding policy priorities that would also merit consideration (even under the headings of S3), I remain convinced that the interpretation of smart specialisation which has shaped the policy research, theory and practice proposed here will emerge as an especially fruitful source of empirically and theoretically grounded economic policy insights – for Europe and beyond. A key question that regional policymakers and stakeholders are facing very often concerns structural transformation: how to transform regional/national economies to shift towards more competitive areas and/or support transitions towards sustainability goals? A first-level answer is about building capacities at various levels of the innovation system: improve human capital, firms’ capabilities, universities’ transfer of knowledge, capital market; in other words, improve everything! This type of answer follows a horizontal logic. There is no preferential intervention in terms of predetermined fields or sectors. This is of course an important logic. A second-level answer involves two steps: i make explicit what the necessary transformations are in which sectors or fields of the regional economy. In any economy, some sectors are more important. They have more opportunities and potential. They thus require preferential interventions. ii improve the chances of microsystems of innovation being formed – specific to the sectors concerned and desired transformations. This second-level answer focuses, therefore, on the deployment of innovative activity and creation of new connections among innovation actors within specific domains, enabling the region concerned to transform its most important structures and develop new competitive

316

DOI: 10.4324/9780429398971-27

Smart specialisation

advantages based on these transformations. To efficiently achieve such transformation, S3 builds on the logics of agglomeration effects in innovation and density of projects. Structural transformations through innovation can lead to various outcomes – including the modernisation of traditional industries, the diversification of such industries towards new emerging markets, the transition of industrial structures to a sustainable trajectory of development and the radical creation of new (sub-)sectors. This second-level answer is complementary and not a substitute for the first level.

24.2

A short history of S3

One question that has been repeatedly addressed for the last 20 years in regional policy discussions is whether there was a better alternative to a policy that spreads R&D investments thinly across several frontier technology and research fields, and consequently fails to make much of an impact in any one area. A more promising strategy can be for regions to identify the domains where new R&D and innovation activities will complement the region’s other productive assets to create future domestic capability and interregional competitive advantage. Such strategy was designated “smart specialisation” by four economists (Paul David, Dominique Foray, Bronwyn Hall and Bart Van Ark) when they were members of the Knowledge for Growth expert group of the European Commission (EC), chaired by Commissioner J. Potocnick (Foray et al., 2009). Based on a fairly general formulation, European regions have embarked on the design and implementation of their own S3s. Within the framework of the EU cohesion policy, a series of conditions were issued (so-called “ex ante conditionalities”), which had to be fulfilled in order for European regions to access European structural funds. Establishing a smart specialisation strategy was one of these conditionalities – which concerned specifically the allocation of structural funding to R&D, innovation and competitiveness. Right from the start of S3 as ex ante conditionality within the framework of EU cohesion/ regional policies, the concept was thus not tight – it was lacking transparency, verifiability and broad consensus. However, the goal of the EC was to go fast and proceed to immediate implementation. The results of this policy were, logically, only partial and imperfect and it is in any case too soon to attempt a final assessment of them.1 However, what is certain is that both academic scholars and policymakers have already acquired an enormous amount of knowledge! They have learned a tremendous number of lessons from these “natural experiments” and progressed in the reflection concerning S3 concepts and practices. They have also made more progress generally regarding the relevant concepts of industrial policy that should be adopted today, not only in the area of regional policies but also for example of mission-oriented policies dedicated to the resolution of grand societal challenges. It is now clear that there is a significant disparity between, on the one hand, recent academic work carried out at a certain level of abstraction that concerns the concept, its rationality, its basic principles and policy design and, on the other, the actual S3 policies implemented. Recent academic work (e.g. Foray 2019b and Foray et al. 2021) has enabled considerable progress to be made, regarding both the concept and its implementation, thanks to the unique opportunity we’ve had to observe real experiences, participate in them and then, based on this, take another look at our initial work. On the other hand, however, the actual practical implementations have not evolved very much and remain strongly marked by the original instructions given by the Commission and its experts. 317

Dominique Foray

We are therefore confronted with a certain academic dynamism and certain inertia when it comes to policies – which is a very natural or logical disparity: while the academic researcher wants to understand why a particular mechanism doesn’t seem to work and is happy to constantly modify the concepts and instruments he gradually creates based on the feedback and information he receives, policymakers logically find it far more difficult to acknowledge that the programme should be modified – the very one that they were only recently offering to their “clients”. One clear goal of this chapter is to fill this gap – clarifying and explaining the theory and providing the consistent tools and methods for effective smart specialisation policymaking.

24.3

Concepts

The second level of answer (above) implicitly involves a geographical concept and a policy design concept.

24.3.1

Economic geography of innovation

In designing a smart specialisation strategy, regions can address a dual problem – that of differentiation and concentration of their innovation capacities – which is generally poorly dealt with by standard research and innovation policies.

24.3.1.1

Differentiation

each region is different with regard to its history, relative specialisations and socioeconomic, geographic, demographic conditions, etc. This means that a region’s current situation is the outcome of a path dependent process of development. If such a process is not understood and managed (through entrepreneurial initiatives and/or public policies), there is a risk of the region becoming locked into obsolete specialisations. In the best case, regional specific assets and structures become the foundation of the economy’s diversification towards new areas of competitive advantages. Such processes are analysed and modelled with the concept of related variety (Boschma and Frenken, 2011). Regional differences imply that each region can be characterised by specific capacities, potentials and opportunities concerning research and innovation – that cannot be fully fulfilled within the framework of undifferentiated policies, which are limited to the provision of aggregate and generic capacities (education, public research infrastructure, finance). Through S3, each region is therefore invited to particularise itself by identifying these new combinations between regional-specific capacities and regional-specific opportunities that should be explored and developed further.

24.3.1.2

Concentration

Once such combinations have been identified, there is a need for some kind of concentration of resources, agglomeration of actors, encouragement of complementary projects and the provision of new specific public goods in order to advance knowledge and innovation in the selected domains. An essential determinant of the productivity of activities dedicated to innovation is scale, critical mass, and a sufficient agglomeration of actors. It is problems of R&D infrastructure indivisibility, markets for specialised inputs (such as skills or services) 318

Smart specialisation Table 24.1 Four Cases of Regional Innovation Policies Concentration of innovation activities Differenciation of innovation activites

Yes

No

Yes

Smart Specialisation Strategies “Another biotech cluster”

“Unrelated/isolated projects” Horizontal Policy

No

and methods of circulating and recombining ideas and knowledge that give large-scale systems – for example, urban centres – an indisputable comparative advantage when it comes to innovation. Thus each region is well advised to possess this critical mass of innovation actors but here a medium-sized region will be unable to obtain them everywhere. Thus, choices must be made and the concentration process should be guided by the logic of differentiation and identification of new combinations between specific capacities and specific opportunities. Differentiation and concentration are complementary, meaning that each of the two policy objectives reinforces the positive effect of the other so they need to be pursued together. This will avoid spending resources unproductively on policies that pursue only one of the objectives while ignoring the other (Table 24.1). This is the case of the so-called “another biotech cluster” policy, which corresponds to a policy of specialisation without differentiation. Regions want to specialise in the same “good thing” even if there is nothing in the region in terms of assets and capacities that could justify such a choice. Such sheep-like behaviour creates a situation where poorly differentiated regions compete for the same resources. As a result, very few regions will be able to build critical mass in the considered domain and compete successfully at the global level. It might also happen that no winner emerges at all because the potential agglomeration economies are going to be dissipated when too many regions compete for the same factors. The mirror of this “concentration without differentiation” policy is “differentiation without concentration”. In this case, the policy is supporting isolated and unrelated R&D projects; that is projects that are quite disconnected from the regional product space so that they will not benefit from positive locational effects (such as intra-regional spillovers and synergies, thick specialised factor markets and specific R&D infrastructures). In such cases, isolated projects are likely to fail or at least to be relocated elsewhere where similar projects are undertaken, and complementary capabilities are available. A logical consequence of a policy aiming at both differentiation and concentration is that choice matters. It is not true to say that because of globalisation and digitalisation, most regions have no choice. Quite the opposite. For regions wishing to engage in international competition based on innovation, there are always a large number of entry points.

24.3.1.3

S3 and the new geography of innovation

In a recent paper, Jaffe and Jones (2015) mention the two forces that are now driving the geography of innovation – the advance of ICTs has allowed people at geographically distant points to interact easily in the production and consumption of new ideas. In contrast to this first force, increased specialisation of human capital or other inputs may encourage further geographic agglomeration. In other words, the primacy of place (e.g. in clusters) may increase 319

Dominique Foray

with time rather than diminish. S3 tries to optimise the balance between these two forces. It aims at levering the first force (the fact that advanced ICTs enable access to capacities and capabilities which are not available in the region) to strengthen the regional innovation place in a specific field or sector.

24.3.2

Policy design: Planning and entrepreneurial discovery process

Any industrial policy or mission-oriented policy or S3 has, explicitly or implicitly, a planning logic. The simple fact of identifying a transformational goal to be accomplished already represents the embryo of a plan. A plan is not bad in itself and is inevitable in a way when a policy is oriented towards a mission! It indicates a direction and encourages the coordination of investments and the initiative of firms within the framework of this direction. But the plan can be bad if it fails to acknowledge and take into account two crucial problems – its inability to control the inherent uncertainty of innovation (Rosenberg 1992) and its inability to know or foresee the specific needs of economic agents with regard to innovation (Hausmann and Rodrik 2006). In acknowledging this dual inability, planners must content themselves with designing an incomplete plan and must put in place a process designed to elicit information about transformational goals, as well as constraints, gaps and needs that need to be resolved in order to make the desired transformation successful. This process is designated entrepreneurial discovery process (EDP): what is “discovered” rather than “planned” is the knowledge of what and how to do it in order to meet the transformational goal. The fundamental point here is the Hayekian argument that the knowledge regarding what to do and how to do it is not obvious. It is knowledge “of time and place”; localised knowledge that is dispersed, decentralised or divided. Above all, the ex ante knowledge is incomplete. There is always potential for discovery and surprise about what to do as concrete processes of exploration are undertaken. The S3 approach is thus marked by a high level of intentionality and strategic focus. But, it is also characterised by a high level of discovery and initiative by the actors of the innovation process. It is this combination of two policy logics – a planning logic and a self-discovery logic – that constitutes its trademark. This nature of S3 was noted by Paul David – one of the concept’s authors – who said that S3 is neither totally top-down nor purely bottom-up. “The S3 approach is about designing an intermediate process aiming to enhance entrepreneurial efforts and coordination within a framework (a strategic priority) structured by the government”.2

24.3.3

S3 involves several discontinuities relative to usual regional innovation policies

i For a long time, too many regions did not think in terms of policy differentiation – just following undifferentiated recommendations of undifferentiated best practices. Such non-differentiation had the adverse effect of encouraging countries and regions to set their sights on doing the same “good things” to foster the same forms of innovation, which in the end proved to be inconsistent and unrelated to the region’s existing assets and potential, and did not provide any comparative advantage (David 2010). S3 recognises that region-specific capacities, problems and opportunities do require policy differentiation. Following this logic of differentiation, regions have a chance to yield results that will be superior to the past policies produced by undifferentiated recommendations of undifferentiated “best policy practices”. 320

Smart specialisation

ii For a long time, innovation policy required that governments concentrate on a few very aggregate capabilities, like education, health, rule of law, generalised infrastructure as well as generic capabilities in terms of innovation (R&D). Such policies ignored the fact that the inputs needed to innovate in a particular sector are highly specific (Hausmann and Rodrik 2006). S3 recognises the need to design policies which accommodate such a highly specific nature of innovation inputs and infrastructures iii The standard policy mindset is still governed by some sort of principal-agent logic: the principal establishes a plan and generates the list of instruments (incentives) for the agents to execute the plan (Rodrik 2004). S3 recognises that there is no such thing as a perfect planner with all the knowledge needed to build ex ante the detailed plan to be followed (because of the first two discontinuities). Hence the centrality of an interactive process, the EDP, which serves to elicit information on business opportunities and constraints and generate policy initiatives in response.

24.4

Fundamentals of S3

24.4.1 Five principles To attain the general objectives of S3, supporting transformation through innovation of certain important domains of the regional economy – five fundamental principles play a vital role: • Encourage regions to drive transformations and thereby build new competitive advantages on the basis of their specific strengths, potentials and opportunities, rather than doing as others do. • Concentrate on specific priorities. This principle has several purposes: • John Enos (1995) made this point very clearly some time ago: prioritisation is a productive and healthy process for the benefit of regions or countries. They “should put more effort into choosing, in detail and for the future, the direction of R&D – on what products, what processes, into what markets”. And Enos explained: if they do not do it, others will do it for them; and more important, the knowledge and experience acquired in choosing the right direction – what we will call “entrepreneurial discovery” – will be very valuable for carrying out the subsequent stage of product and process innovations. • Second, this principle aims to generate a certain density of actors and activities that are related as they are dedicated to the same priority – an imperative condition to benefit from the resulting synergies, complementarity and agglomeration, which are essential determinants of innovation, creativity and R&D productivity. Concentration achieves increased density. • Third, this is also an important condition for a government to be able to reach the level of input specificity required to innovate in a given industrial or technological domain. This has been a constant argument by Hausmann and Rodrik (2006) – that “the public inputs that innovators require tend to be highly specific in the area in question. There are really very few truly generic inputs for innovation”. But governments cannot address all specific innovation infrastructures and specific services for all markets and activities. Government capacities, both in terms of information (what does each industry need in terms of specific inputs?) and resources (can we afford the provision of all industryspecific public inputs for all sectors?), are indeed limited. They need to choose the areas on which to concentrate. As asserted by Hausmann and Rodrik (ibid), “it is not that choices are desirable, they are simply inevitable”. 321

Dominique Foray

• Concentrate not on structures (for example, do not choose the region’s three most important industries) but on the transformation of these structures. The idea here is that what needs to be supported is not a specific sector per se but the new activities that will contribute to its transformation. This principle has one main purpose, which is to allow preferential interventions, while minimising distortions: it is not enough to be part of a targeted structure (one particular industry) to be helped. It is also necessary to be involved in the desired transformation process. Hence, each priority area includes one or several sectors as well as a direction of change. If both elements are combined and sufficiently well defined to create the density effects mentioned above, they build a priority area, a cornerstone of a smart specialisation strategy. • Favour a logic of bottom-up and decentralised discovery, which means simply that the targeted transformation process will not follow a path that is decided from the top, but will be discovered as the process unfolds. There is therefore no ex ante plan; the “plan” will only emerge ex post as a result of the process. The importance of this principle is related to the recognition that no one government can acquire innate wisdom or the ex ante knowledge about the path to be followed, once a priority area, including the desired transformative directions, has been selected. The EDP logic is activated at various steps of S3: first to identify specific and detailed transformational goals within a broad strategic field; and second, to reveal areas of desirable interventions (called transformative activities) in order to push forward the transformational goals. • Implement a continuous process of EDP. EDP cannot be limited to a one-off exercise, but should be a continuous element throughout the strategy implementation. Indeed the implementation of the strategy – including several programmes and actions – is characterised by a high level of uncertainty regarding how each of these actions will evolve. Launching the various activities is like starting a voyage of discovery – to use Hirschman’s expression (2015). By definition, discoveries involve success, failures and surprises and it is critical for smart policy development to include feedback mechanisms, monitoring principles and flexibility to maximise the informational effects and spillovers of all discoveries. A new way of approaching project funding (Rammer and Klingebiel 2012) is very much suited to this objective of keeping the road map as flexible as possible: instead of one single financing decision, made at the start of the project, the authors elaborate a multiple and sequential decision model that allows projects that are not working to be discontinued sooner and the volume of financing allocated to those that are progressing to be increased.

24.4.2 A simple process A key operation of S3 is to put in place a process of strategic interactions between the government and stakeholders to discover i) specific transformational goals within broad fields; ii) the gaps, problems and opportunities which characterise such transformations; and eventually iii) the policy initiatives to be taken in response. An S3 process thus involves two stages of prioritisation and one stage for building a road map within each priority area. The two stages of prioritisation allow the process to evolve from the identification of broad strategic fields and general directions of changes to more specific priority areas and transformational goals. The outcome of this three-stage process (Figure 24.1) is a transformational road map for each priority area. A road map includes a collection of transformative activities – each of 322

Smart specialisation 1

2

A strategic field and a transformation direction

3

Priority area and transformational goal 1

Transformational roadmap 1 – a collection of transformative activities (problems, opportunities and policy responses)

Priority area and transformational goal 2

Transformational roadmap 2 – a collection of transformative activities (problems, opportunities and policy responses) feedback

Planning…

Figure 24.1

…Entrepreneurial discovery..

à

S3 as three-stage process.

these activities associates the identification of a problem or an opportunity with a policy intervention. This collection of transformative activities thus forms the basis for designing and implementing policy initiatives (such as call for proposals, procurement, prizes, the provision of new infrastructures or new services, etc.). Feedback is important – particularly from stage 3 to stage 2: if the road map is assessed as not “rich” enough (small number of activities, insufficient capacities in the current system to undertake the activities, etc.), the definition of the corresponding priority area at stage 2 was probably not mature enough or not well defined and therefore needs to be revisited. This process involves the two logics of planning and entrepreneurial discovery mentioned above (section 24.3.2). The planning logic is predominant at stage one and then decreases in intensity while the EDP logic is not present at stage one, but starts to be activated at stage 2 and increases in intensity as the process unfolds. Box 24.1 describes a particular case of S3 – undertaken in the region of Skåne.

Box 24.1

Skåne region – A concrete illustration of the three-stage S3 process

Explanations – The Region Skåne (Sweden) has put in place a process to design and implement its S3 which follows quite well the logic described in this chapter. At a first stage, large priority areas have been identified (column 1). At this stage, the logic of centralised and top down decisions is dominating, even if all relevant stakeholders were involved in the decision process. At a second stage, each of the three large priority areas were sub-divided into more specific transformational goals. For instance, within the large priority area of innovation for a sustainable food system, seven sub-goals were identified (see some of the seven sub-goals in the column 2). It is obvious that this second stage is very important. Breaking the broad priority areas into more specific transformational goals is key to enhance coordination between decentralised agents (in particular firms, research partners and public agencies), to provide the specific collective goods (infrastructures, services) and increase relational density. At this stage, the planning logic is complemented by an entrepreneurial discovery logic. Indeed, the level of specificity which is appropriate to

323

Dominique Foray

define the transformational goals makes entrepreneurs and other stakeholders central in eliciting information and knowledge about the specific capacities and opportunities of their sector (here the agro-food industry). Finally, at a third stage, the entrepreneurial discovery process is activated again to identify problems and gaps which need to be addressed to reach each transformational goal and the tailored policy initiatives in response. For example, within the framework of the transformational goal –Realising unique opportunities for knowledge-based innovative product and processes development in the food and packaging sector – the activities described in column 3 are now undertaken. This collection of activities is the transformational road map. The Skåne example provides an excellent illustration of the right way of thinking of the logic of S3: a discovery process – one where firms, researchers and the government learn about constraints, problems and opportunities within a very specific area and engage in strategic coordination to generate unique and taylored policy initiatives in response.

1 – Broad strategic fields

Innovation for sustainable food

Life science & health

2 - Priority areas and transf. goals Recycling plastics

3 -Roadmaps

Pilot to implement new techniques in food contexts

Public meals innovation

Public-private partnerships for R&D in food process

Process development in food and packaging

Tech sector Cultivation and production of plant-based foods

Training programmes

Industrial doctoral projects

Source – Tillväxtverket – Innovation for a sustainable food system – Skane – draft paper-2021

324

Smart specialisation

24.4.3

EDP process and outcomes: Transformational road maps and transformative activities

As mentioned, the EDP needs to operate at two stages of the process (Figure 24.1). At stage 2, the goal of the EDP is to explore a broad strategic field in order to highlight more specific and narrow priority areas – each being defined as the association of a specific area within the strategic field with a transformational goal. At this stage, we will talk of an EDP of low intensity because top-down insights and decisions still matter. The mutual influences of planning and EDP are well balanced. Once these more specific priority areas have been defined, the next goal of EDP is to elicit information about the problems, needs and opportunities that are characterising a given priority area and to propose the appropriate policy initiatives. The EDP is therefore a unique mechanism to build the transformational road map. It is unique and unavoidable because – as already stated – what needs to be done to meet a specific transformation goal cannot be planned and predicted from the top. It needs to be “discovered” and revealed by practitioners, researchers and other stakeholders. This bottom-up discovery process will uncover a collection of complementary activities (all being undertaken and pertinent in relation to the transformation’s objective) – covering a multitude of dimensions. This road map could under no circumstances have been imagined or predicted by the government. To summarise the role of EDP, there is certainly a planner (to drive stage 1 and support stages 2 and 3) but not an omniscient planner. As the process shifts towards stages 2 and 3, the planner takes back seat to experimentalist governance (Morgan and Marques 2019). It is for the governance at stages 2 and 3 that the following quotation makes a lot of sense: “What if, as I and many others assume, there are no principals with the robust and panoramic knowledge needed for this directive role” (Sabel 2004). The transformational road map and transformative activities it comprises are therefore the essential outcome of the EDP. As illustrated in Box 24.1, the transformative activities (column 3) are not just about R&D but cover many topics and issues, which represent critical steps towards the desired transformation, possibly including human capital formation, R&D infrastructure development, technology diffusion and adoption, network generation, etc. The fact that there are many issues to be addressed just reflects that innovation and transformation have multiple determinants (not only R&D) (see below, 44). A transformational road map has three key properties that need to be well understood and exploited so that it can serve as a catalyst for collective action by firms, suppliers, research partners and customers towards the desired structural transformation.

24.4.3.1 Strategic complementarities Determinants of productivity/innovation are multiple and the complementarities among them are key. Obvious examples involve the complementarity between the invention of generic technologies (or GPT), the development of applications for several sectors and the adoption of these applications within the concerned sectors; or the complementarity between the investments in some specific R&D fields and the investments in the corresponding specialised scientific and engineering skills and capabilities; or the complementarity between the support of SMEs’ competitiveness and the provision of technological infrastructures. When properly managed, such strategic complementarities among activities can stimulate the emergence of a persistent pattern of change (Milgrom and Roberts 1990).

325

Dominique Foray

A transformational road map is designed to capture all these complementarities and demonstrates that there is great advantage to be gained in launching activities simultaneously, resulting in even stronger transformative activity.

24.4.3.2

Spillovers

The second property concerns the notion that all activities that are part of the transformational road map have the potential to provide knowledge and information spillovers. When the initial projects and experiments undertaken within an activity are known to be successful, other agents are induced to join the transformative activity. According to Hirshleifer (1971), public information regarding project successes, failures and surprises has high social value in redirecting productive and investment decisions. Thus, the governance of the transformational road map – once the activities have been defined and start to grow – should be done in such a way as to maximise the spillovers to all the stakeholders – including potential entrants.

24.4.3.3 Aggregation level One difficulty is to place the S3 operations at the right level of aggregation. It is quite obvious from the third principle mentioned in section 24.3.1 that S3 should not be a process embracing the whole sector – it is not a sectoral policy. Rather, S3 is about transformation and it is obvious that not all firms in one sector will be committed to the desired transformation. On the other hand, it should not be a process selecting individual projects while disregarding relatedness and coordination among projects – because such a policy will fail in generating the necessary relational density and agglomeration of actors. Between these two levels – a sector as a whole and individual, isolated projects – a collection of complementary activities can be envisaged, all involved in the same transformation process: a transformational road map. This intermediate level of aggregation is a key principle of a new generation of industrial policies. It invites us to make the key objectives of S3 explicit: focussing on transformations; building systems of complementarities (or systems of innovation) to attain them.

24.4.4

S3 aims at boosting both vitality and inclusion

It is not easy for regional policy to promote both vitality and inclusion. Policies aimed at promoting rocket science and high-tech entrepreneurship will possibly have an inclusive effect in the long term thanks to the fairly usual macroeconomic sequences – as described by Ned Phelps (2006) in his Nobel Prize lecture – or because of the potential effects of innovation on social mobility so well described by Aghion and Akcigit (2015). However, generally speaking, such positive effects on inclusion will only be realised in the long term while in the short term, such policies are essentially discriminating and exclusive and will to a great extent benefit talented students from a few top campuses – assisted and supervised by very selective financial actors. However, a region needs to promote both vitality and inclusion also in the short term. The question, therefore, is how to provide a regional economy with the right mechanism to boost both vitality and inclusion. We believe that S3 belongs to this class of precious mechanisms as it is illustrated with the case described in Box 24.2. 326

Smart specialisation

Box 24.2

Lombardy S3 – promoting both vitality and inclusion

At the headquarters of the Lombardy region, innovation policymakers are facing such a problem (that any region is facing to some extent). Do we want to just focus on dynamism (a start-up policy which hopefully will generate some long-term effects on inclusion) or do we want to design a policy that could generate at the same time – that is today - dynamism and inclusion? Finlombarda SpA, Lombardy financial and innovation agency, supports a bunch of great start-ups – inventing new high-tech products and services with strong application potentials in the agrifood sector. Based on a high-tech policy only, the entrepreneurial activities are going to be stimulated and this will be beneficial to a small part of the Lombardy economy – a few indicators will improve and these are not the worst ones (patent, VC attractions, highly skilled jobs) – but the inclusion effect will be negligible. The point here is to involve the agrifood sector as a huge reservoir of potential adopters of these new technologies. The challenges are multiple: addressing human capital and capability problems, fixing the adoption externalities, addressing coordination failures and providing some specific public goods. The whole policy is probably much more difficult to design and implement – it will involve different kinds of actors (such as vocational education institutions; specialised services and platforms, clusters) and will have to address many barriers and obstacles to innovation diffusion in traditional sectors. The choice is, therefore, between helping a few nice guys with brilliant ideas or undertaking the proper actions to support a real transformation of certain structures of the economy. And this is what the idea of smart specialisation tries to suggest: shifting from “just” a high-tech policy to a policy aiming at supporting the development of a real transformational process which would be likely to drive structural changes – not only in high tech but in the huge agrifood sector.

The lessons from this case are obvious. The modernisation of the agrifood sector involved two goals: encouraging young innovative firms by equipping their ecosystem with all the complementary capabilities needed AND addressing the innovational complementarities between the high-tech and traditional sectors. A policy designed to support such transformative activities would entail the provision of innovation services and infrastructures, formation of new human capital, subsidisation of technology adoption on top of helping and cherishing the start-ups and their ecosystem. For many regions, as in the case above, priority areas are about transforming, modernising or diversifying the existing industries. The goal is therefore to include some existing structures and capacities in an innovation strategy. Then, as the transformational road map is developed, the point is not to invent at the frontier but rather generate innovation complementarities in existing sectors. Such innovation-related activities might include building up human capital, adopting (not inventing) new technologies, diffusing novel management practices or generating complementarities between key enabling technologies and traditional sectors. As Trajtenberg (2010) wrote regarding innovation: “there is not only one game in town”. The fact that such innovation-related actions ultimately represent the key to economy-wide growth in most regional economies needs to be reflected in the choice of the relevant priority areas and development of the associated transformational road maps for a given region. 327

Dominique Foray

A policy to promote both dynamism and inclusion is not just a policy that would support pushing more resources into the economy (more research infrastructure, more human capital), because these resources will ultimately be largely captured by the top science/hightech ecosystem but instead, it aims also at pulling some existing resources (of the traditional sectors) into innovation activities.3 In this sense, smart specialisation has an inclusive component because the strategic domains and transformative activities that are identified and selected are not limited a priori to a certain (high-tech) part of the economy. Indeed, in many cases, they will be parts of “old” industries in a declining structural change, or they belong to the category of industries that are already successfully growing and competitive but with potential for even more advances. This is the raison d’être of smart specialisation: new combinations between existing capacities and new opportunities can emerge everywhere in the regional economy.

24.4.5

The 5 Ds

The whole transition process from a broad strategic field (stage 1) to more specific priority areas and transformational road maps (stages 2 and 3) is a key transition. If properly done – thanks in particular to a robust and transparent EDP process – it enables the good outcomes of an S3 approach to be achieved, designed to transform the structures of the regional economy, and which I have grouped under the heading of the 5 Ds. Direction of change Relational Density Regional Differentiation Entrepreneurial Discovery Distributed capacities We can explain the significance of each of the 5 Ds in more detail as a desirable property of a transformational road map. Once properly defined, the transformational road map • … concretises a certain direction of change, initially expressed by the specific priority area and transformational goal, and reveals some initial guidelines concerning the course of action to achieve this change; • … enables the transition from broad strategic fields and general direction, which are to a large extent similar from one region to another, to a deep regional differentiation. In fact, similar priorities across regions will lead to different transformational road maps, as the latter are designed as a specific response to problems and opportunities specific to the particular region; • … creates relational density and increases the chance of reaping the benefits of a certain coordination between the projects and actors involved in this transformation. This is due to the fact that all activities are related because they all contribute in one way or another to the same structural transformation in the priority area; • … covers a large number of issues, including of course R&D but also the formation of human capital, corporate management, adoption of new technologies, etc. It is therefore a collection of distributed capacities and projects (instead of a single major project such as the creation of a new specialised R&D institute, frequently destined to become the proverbial white elephant); 328

Smart specialisation

• … is the preferred framework for entrepreneurial discovery. At the end of the priority definition phase, it is impossible to know what the outlines and content of the transformative activities will be. They are built and developed on the basis of the entrepreneurial discovery process.

24.5

Conclusion

Recent and current experiences in various regional settings – undertaken by the author of this paper and his colleagues from BAKS34 – show clearly that S3 design and implementation are feasible but challenging. It is difficult to identify a relevant priority area, but even more so to translate it into more specific transformational goals and finally move towards transformational road maps and activities. The main problem is that the construction of the transformative activity requires a very detailed understanding of the coordination relationships between different types of investments from the policy agency as well as a deep knowledge of the specific inputs, which are needed to innovate within a particular set of industries. These information requirements are hard to meet and place the S3 approach into a “haute couture” logic (as opposed to the “ready-to-wear” logic). “Haute couture” is by definition costly because nothing is really replicable from one S3 to another. Each S3 approach needs to build its own transformational road maps, given the specific constraints, capacities and opportunities in the considered region and sector. Of course, the logic of entrepreneurial discovery should be considered here as a partial solution to the information problem. This is the role of stakeholders – firms, research, etc. – to discover what needs to be done; the kind of investments required at a very high level of details and specificities. The notion of entrepreneurial discovery should not – by any means – be viewed as just an elegant academic trick – theoretically useful to minimise the top-down logic of the process. It is a true necessity – to overcome the informational challenge raised by the logic of S3 at steps 2 and 3. However, S3 is in any case a costly approach. It is difficult to handle. Impact evaluation will be tricky as well. Generating strong evidence about a causal effect of an S3 approach on the economic situation of the region concerned during the subsequent period of time is practically impossible because a specific S3 is a highly complex instrument with no simple treatment effect and no obvious counterfactuals. We do think, however, that the implementation concept described in this paper has a great future! It could prove helpful not only in the context of regional policies, but also in the area of the mission-oriented innovation policies that are discussed these days to address various grand and global challenges (Foray 2018a and b). For all such policies, there is always the problem of enhancing entrepreneurial efforts and coordination within a framework, which is structured from the top. For all these policies, the tension between topdown prioritisation and bottom-up decentralised actions has to be managed via an efficient and effective policy design (Foray 2019a). For all such policies it makes no sense to hide the planning component, which is part of the very foundation of mission-oriented and S3 policies, just because planning is not a wellreceived word today. It makes no sense to deny the fact that there is indeed a planning logic since this is the very nature of S3 to define priorities and transformation targets, which in turn will determine preferential interventions. S3 involves a high degree of intentionality – this is one of its key positive characteristics. However, the planning logic of S3 is not the kind of planning that ignores the existence of uncertainty (Hirschman 2015). Planning and 329

Dominique Foray

bottom-up discoveries are the two inseparable logics of a policy that is characterised by a high level of intentionality and strategic focus, while recognising both uncertainty and the inability of the planner to decide on the transformational road maps and predict a project’s success or failure. This provides regional governments and public agencies with an appropriate toolbox to manage difficult transitions, such as getting out of what today appear as obsolete specialisations (e.g. fossil-based technologies and sectors). It is the main contribution of this paper to show how the first period of S3 implementation in the EU and beyond and the feedback and learning processes derived from this unique policy experiment allow us to better understand what kind of policy design can be effective to generate structural transitions and strategic initiatives. The economics of smart specialisation policy is thus a good narrative for policymakers at both national and global levels. The message is that with such new policy instruments, policymakers are at the forefront of a new generation of industrial policy – which has multiple applications, of course in regional policy (Morgan and Marques 2019, Foray et al. 2021), development and industrial policy (Hausmann and Rodrik 2006, Sabel 2004 and Rodrik 2004), mission-oriented policy (Foray et al. 2012) and high-tech policy (see Azoulay et al. 2018, on the US ARPA family of programmes). In observing these different applications, I would like, when all is said and done, to suggest the idea of a collective evolution towards a new vision of industrial policies: that of the combination of two policy logics – a planning logic and a discovery logic – that constitutes its trademark. It should be noted that these two policy logics are frequently opposed in the literature as well as in practice. The future of these policies is to combine them thanks to robust and efficient policy designs. Our work on S3 contributes to this direction of evolutionary policy research.

Notes 1 Promising evaluation works have been achieved most recently. See for instance Rigby et al. (2022) as well as Prognos and CSIL (2021). 2 Paul A. David, personal communication, 2012. 3 An example of a regional policy which emphasises the push logic is given by Gruber and Johnson in their book Jum-Starting America (2019). 4 BAK S3 is a spinoff of BAKeconomics (Basel, Switzerland) which develops methodologies and supports regional gouvernments for the design and implementation of smart specialisation strategies.

References Aghion P. and Akcigit U (2015) Innovation and Growth:the Schumpeterian perspective. Azoulay P., Fuchs E., Goldstein A. and Kearney M. (2018) Funding breakthrough research: promises and challenges of the ARPA model, Innovation policy and the economy, Vol. 19, J. Lerner and S. Stern eds. Chicago: University of Chicago Press. Boschma R. and K. Frenken. (2011) Technological relatedness and regional branching, in Dynamic geographies of knowledge creation and innovation, H. Bathelt, M.P. Feldman and D.F. Kogler eds. London: Routledge. David P.A. (2010) Comments on enhancing Bulgaria’s competitiveness and export performance through technology absorption and innovation. Washington DC: World Bank. Enos J. (1995) In pursuit of science and technology in Sub-Saharan Africa. London: Routledge. Foray D (2018a) Smart specialization as a case of mission-oriented policy – a case study on the emergence of new policy practices, Industry and Corporate Changes, 27(5), special issue on Mission oriented innovation policy.

330

Smart specialisation Foray D (2018b) Smart specialization strategies and industrial modernization in European regions – theory and practice, Cambridge Journal of Economics, 42(6), special issue – The dynamics of industrial and economic renewal in mature economies. Foray D (2019a) On sector-non-neutral innovation policy: towards new design principles. Journal of Evolutionary Economics, 29(5), special issue on Evolutionary Innovation Policy. Foray D (2019b) In response to Six critical questions about smart specialization, European Planning Studies. Foray D., David P.A. and Hall B. (2009) Smart specialisation: the concept, Knowledge for growth. Brussels: European Commission. Foray D., Eichler M. and Keller M. (2021) Smart specialization strategies – insights gained from a unique European policy experiment on innovation and industrial policy design, Review of Evolutionary Political Economy, 2, 83–103. Foray D., Mowery D. and Nelson R. (2012) The need for a new generation of policy instruments to respond to Grand Challenges, Research Policy, 41(10). Gruber J. and Johnson S. (2019) Jump-starting America. New York: Public Affairs Hausmann R. and Rodrik D. (2006) Doomed to choose, working paper, Cambridge, MA: Dept. of Economics, Harvard University. Hirshleifer J. (1971) The private and social value of information and the reward to inventive activity, American Economic Review, 61, 4. Hirschman A.O. (2015) Development projects observed. Washington DC: A Brookings Classic. Jaffe A. and Jones B. (2015) Introduction in The changing frontier. NBER, The University of Chicago Press. Milgrom P. and Roberts J. (1990) The economics of modern manufacturing: technology, strategy, and organization, American Economic Review, 80(3). Morgan K. and Marques P. (2019) The public animateur: mission-led innovation and the ‘smart state’ in Europe, Cambridge journal of regions. Economy and Society, March. Phelps E (2006) Macroeconomics for a modern economy. Prize Lecture. Prognos and CSIL (2021) Study on prioritisation in S3 in the EU. European Commission. Rammer C. and Klingebiel R. (2012) Public funding of innovation projects: is it time for a more flexible approach?, ZEW Policy Brief, 2. Mannheim: Center for European Economic Research. Rigby D., Roesler C., Kogler D., Boschma R. and Balland P.A. (2022) “DO EU regions benefit from Smart Specialisation principles?, Regional Studies, 10.1080/00343404.2022.2032628 Rodrik D. (2004) Industrial policy for the twenty-first century, CEPR discussion paper, n°4767. Rosenberg N. (1992) Economic Experiments, Industrial and Corporate Change, 1(1), 181–203 Sabel C. (2004) Beyond principal-agent governance: experimentalist organizations, learning and accountability. New York: Columbia University Press. Trajtenberg M. (2010) Development policy: an overview, D. Foray ed. The new economics of technology policy. Cheltenham: Edward Elgar.

331

25 EVOLUTIONARY ECONOMIC GEOGRAPHY AND POLICY Ron Boschma

25.1

Introduction

In a recent overview on evolutionary economics by its leading proponents (Nelson et al. 2018), the geographical dimension of economic evolution was almost completely overlooked. However, economic evolution can only be understood when not only time but also the dimension of space is fully incorporated. Economic evolution is fundamentally uneven in space, showing persistent inequalities at multiple spatial scales. This neglect of the geographical dimension in evolutionary economics is remarkable given the fact that a large body of literature on evolutionary economic geography (EEG) has developed since the 1990s. The theoretical foundations of EEG have been laid down in many publications (e.g. Boschma and Lambooy 1999; Boschma and Frenken 2006; Martin and Sunley 2006; Boschma and Martin 2010), and these foundations continue to be explored and discussed (see e.g. MacKinnon et al. 2009; Boschma et al. 2017; Henning 2019). In the vanguard of this, a massive body of empirical studies has been published on the geographies of firms, technologies, science, innovations, jobs, industries, clusters, networks and their evolution over time (Boschma and Frenken 2018). Evolutionary economics has been heavily engaged in debates on public policy (Edler and Fagerberg 2017), the future of industrial policy (Rodrik 2004), the changing role of the state (Mazzucato 2013), the need for mission-oriented policies (Mazzucato 2018), how to promote transformative change (Schot and Steinmuller 2018), and how to tackle societal challenges (Kuhlman and Rip 2018). In these policy accounts, geographical aspects of economic evolution have not been given full consideration. In evolutionary economic geography instead, interest in fundamental policy debates has been rather implicit. This policy interest is growing though, especially with regard to the smart specialization policy in the EU (Foray 2015; Balland et al. 2019). This chapter aims to shed light on what EEG might have to offer in terms of these policy debates, especially with respect to the place-based dimension of innovation policy (Barca 2009). This will be illustrated by taking the example of the smart specialization policy (S3) which represents a form of place-based innovation policy in the European Union in which

332

DOI: 10.4324/9780429398971-28

Evolutionary economic geography and policy

evolutionary concepts like directionality and relatedness have been widely adopted since its start in 2014 (Foray et al. 2009; McCann and Ortega-Argilés 2015).

25.2 EEG and smart specialization policy The smart specialization policy reflects a vision on regional innovation that is embedded in place-based capabilities. The core idea is to prioritize new domains of specialization in regions that complement and leverage their local capabilities (Foray et al. 2009). The point of departure is that region-specific capabilities define the set of opportunities for developing new growth trajectories in regions. Regions should refrain from targeting new domains in which they lack relevant capabilities. Instead, they should go for domains that build on local related variety (Frenken et al. 2007) and promote the diffusion of knowledge across domains, because such cross-fertilizations lay the foundations of new growth paths. The policy objective is to identify strong capabilities in regions and use them to move into new activities that exploit local synergies and avoid competition with other regions. This implies a rejection of one-size-fits-all policy, in favor of a place-based policy, that is attuned to the capabilities and demands of regions (Tödtling and Trippl 2005; Barca 2009). The concept of relatedness has been used as a key principle in the EU guidelines for S3 to select future domains in regions (Foray et al. 2012; McCann and Ortega-Argilés 2015; Iacobucci and Guzzini 2016). It builds on evolutionary concepts such as bounded rationality, local search, proximities, networks and path dependence to understand regional development. Relatedness is a principle that identifies activities that demand similar and complementary capabilities (Teece et al. 1994, 1997; Breschi et al. 2003). It has become a key factor to explain regional diversification processes in which the dimensions of time and space are being combined (Boschma 2017; Hidalgo et al. 2018; He and Zhu 2019). That is, regions tend to diversify into new activities that are closely related to their existing capabilities. Place-specific capabilities condition diversification in which geographical, cognitive, social and institutional proximities enable the transfer of capabilities from existing to new activities. Empirically, there is overwhelming evidence that related diversification is far more common than unrelated diversification in regions, no matter whether it concerns diversification in new industries (Neffke et al. 2011), jobs (Muneepeerakul et al. 2013) or technologies (Rigby 2015). Balland et al. (2019) developed a framework for the smart specialization policy along the two dimensions of relatedness and complexity. Relatedness refers to the costs of moving into a new activity. These costs will be lower the higher the overlap between the required capabilities of the new activity on the one hand, and the supply of existing capabilities in the region on the other hand. The more related they are, the less risky and less costly it is to develop this new activity. Complexity refers to the potential economic benefits of diversification. The benefits will be higher the more complex activities are (Hidalgo and Hausmann 2009). As complex activities combine many capabilities, it is harder for other regions to copy and develop them. This makes that complex activities may provide a sustainable source of regional competitiveness (Maskell and Malmberg 1999; Fleming and Sorenson 2001). Low-complex activities can be mastered and produced by many regions instead, which implies their economic value tends to be lower (Balland and Rigby 2017; Antonelli et al. 2022). This policy framework enables to identify and map potential activities in which a region does not possess a relative advantage by calculating their degree of relatedness and their 333

Ron Boschma

complexity relative to the existing knowledge stock of the region. This results in a distinction between different policy strategies that represent different risk-return profiles (Balland et al. 2019). One policy strategy targets activities that promise above-average returns (high complexity) at relatively low risk (high relatedness), as it builds on existing capabilities in the region. This may be an attractive policy option especially for advanced regions, as their opportunity space allows them to diversify into high-complex activities that can exploit relevant (related) local capabilities (Pinheiro et al. 2022b). Rigby et al. (2022) assessed whether European cities that follow the logic of this policy (i.e. developing new technologies that increase the complexity level of their economy and that are closely related to local capabilities) showed higher economic performance in the period 1981–2015. They found that cities diversifying into related and complex technologies (while exiting less related and low-complex technologies), enjoyed higher GDP growth than cities that did not. Davies and Maré (2021) found that relatedness and complexity promoted employment growth in the largest cities in New Zealand. Peripheral regions find themselves in a very different situation (McCann and OrtegaArgilés 2015; Morgan 2015; Foray 2019). These lagging regions lack potential activities for future development that score high on relatedness and complexity simultaneously (Pinheiro et al. 2022b). For them, a more viable policy strategy is to target potential activities that score high on relatedness but low on complexity. This reflects a relatively low-risk strategy: it will provide new job opportunities that are likely to match relatively well with the local supply of skills, because it builds on related capabilities. However, expected benefits might be relatively modest, because such policy strategy targets low-complex activities that are likely to be exposed to strong competitive pressure from other regions. To break out of such a low-complexity trap, these regions might consider another type of policy outlined by Balland et al. (2019) that targets activities that are far removed from the existing knowledge base of the region. This involves a high-risk strategy that requires strong and massive policy intervention (Alshamsi et al. 2018; Hartmann et al. 2020). While the chances of success might be low because their focus is on developing something brand new and complex, when successful, the complexity level of the economy of the region will increase, yielding high benefits. This policy strategy is likely to be very difficult and highly risky though. Countries like Brazil and Russia (Hartmann et al. 2020) and regions in the United States and the EU (Pinheiro et al. 2022b) have shown how difficult it is to escape from such a low-complexity trap.

25.3

Design and implementation of the smart specialization policy

So far, we have focused on directionality and priority setting of S3 based on local capabilities and complexity. The smart specialization policy has also attached great importance to the design of the policy process, also known as the entrepreneurial discovery process (Foray et al. 2011). Governance issues and the role of collective agents have been incorporated in smart specialization (Aranguren et al. 2019). Still, scholars have raised concerns about the actual implementation of S3, and how this differs across regions (Matti et al. 2017; Fratesi et al. 2021). Entrepreneurial discovery stands for a bottom-up policy process in which a diversity of local stakeholders are involved to discuss, assess and select promising activities for future development (Rodrik 2004; Foray et al. 2011; Foray 2015). It has some resemblance with 334

Evolutionary economic geography and policy

the regional innovation system concept that focuses on a wide array of local actors that do, or do not, interact and engage in collective action, depending on governing institutions (Asheim et al. 2019). What S3 adds to this regional innovation system perspective is the explicit role of relevant capabilities and the directionality of priority-setting and policy. Studies have looked into this process of priority-setting in the smart specialization strategies in the EU and have assessed the extent to which targets were actually regionspecific, as the S3 guidelines prescribe (McCann and Ortega-Argilés 2016; Pagliacci et al. 2020). What they conclude is that S3 strategies do indeed make choices to a greater or lesser extent, but these priorities are often quite broadly defined. Regions also tend to focus on different priorities which reflect their local capabilities at least to some extent (D’Adda et al. 2019; Trippl et al. 2020; Deegan et al. 2021). Other studies are more critical (Iacobucci and Guzzini 2016; Di Cataldo et al. 2022). Marrocu et al. (2022) concluded that S3 strategies in European regions show limited potential to develop new growth paths that leverage local capabilities. However, this regional specificity of the smart specialization policy is also important for another reason, besides the need to focus on local capabilities per se (Boschma 2022). There is increasing evidence that opportunity spaces of regions look very different, which call for different types of innovation policy (Tödtling and Trippl 2005). Core urban regions tend to have many opportunities to move into complex activities because they can build on relevant (related) capabilities to do so. This still requires strong policy intervention though, despite regular claims that related diversification can do without (Grillitsch et al. 2018). Market and system failures (Hausmann and Rodrik 2003; Frenken 2017) need to be tackled through the public support of entrepreneurship, educational reforms, research capacity-building and institutional change, to ensure these local opportunities are exploited. There are many factors that can make regions fail to diversify into related activities, such as laws and regulations that discourage the mobility of entrepreneurs and workers from related industries (see e.g. Klepper 2010), a poor entrepreneurial culture (Fritsch and Wyrwich 2014), restrictive social norms (De Vaan et al. 2019), weak university-industry linkages, and a lack of (risk) capital investment (Florida and Kenney 1988), among other factors. It comes as no surprise, therefore, that many diversification potentials in regions are not activated in practice. Empirical studies show indeed that the amount of potential entries in related activities in regions far exceeds the actual number of entries in related activities in regions (Boschma 2017). At the same time, core urban regions also have the most advanced research infrastructure, a rich supply of human capital, unrelated variety, and international connectivity, to move the technological frontier and diversify in unrelated activities (Castaldi et al. 2015; Zhu et al. 2017; Pinheiro et al. 2022a). But above all, unrelated diversification requires the build-up of completely new capabilities, in terms of knowledge, skills and instititions (Neffke et al. 2018). This requires collective action and experimentation in which policy is bound to play a major role, due to fundamental uncertainty (Hausmann and Rodrik 2003) and the presence of transformational failures (Schot and Steinmuller 2018). One example is policy intervention that aims to bridge the cognitive distance inherent in unrelated combinations that would remain unexploited otherwise (Janssen and Frenken 2019). The case of old industrial regions is very different. These concern a large group of regions in the Rustbelt of the United States and Europe that used to belong to the most technologically-advanced regions but now find themselves trapped in mature trajectories (Grabher 1993; Hassink 2005; Evenhuis 2016). Old industrial regions tend to show the 335

Ron Boschma

highest diversification potential in simple activities, not in complex activities (Balland et al. 2019). Their industrial history makes it hard for them to move into more complex activities through related diversification. One way out of such a low-complexity trap is promoting unrelated diversification, but this remains a risky and bumpy road, full of obstacles. One major obstacle mentioned in the context of the middle-income trap is that countries might have a weak ability to pursue effective public policies and induce institutional change (Lee 2013; Doner and Schneider 2016; Aghion and Bircan 2017). Instead, the policy focus on related diversification could be a less risky alternative to transform old industrial regions and increase their complexity. There is a growing awareness that radical transformations do not necessarily occur through unrelated diversification only. For instance, technological breakthroughs primarily make related combinations, rather than unrelated combinations (Boschma et al. 2023). And there is increasing evidence that green diversification in regions is enhanced by local capabilities from related activities, even from so-called ‘dirty’ activities (Tanner 2016; Van den Berge et al. 2020). In other words, radical change may rely on relatedness to a considerable degree. The case of peripheral regions is yet another one. Low-income regions have a greater tendency to diversify in activities related to their own activities than high-income regions (Petralia et al. 2017; Xiao et al. 2018). At the same time, lagging regions run the risk of being trapped in a low-complexity state. Pinheiro et al. (2022b) indeed observed that low-income regions have diversification potentials primarily in low-complex activities. To escape from this low complexity trap requires a serious and concerted policy effort. However, there is high risk of policy failure where there is little experience to build on. Duplication of policy efforts is also likely to happen when priorities are not embedded in the regional context, and when ‘missions’ and ‘grand societal challenges’ become popular drivers of regional innovation policy (Mazzucato 2018). Such policy is also likely to build cathedrals in the desert. This is a typical policy failure that happened more often than not when industries were encouraged to locate in peripheral regions where they developed little interaction with their surrounding economy, without any significant positive spillovers, because local firms lacked the absorptive capacity and local institutions were generally weak (Rodríguez-Pose and Wilkie 2015). According to Foray (2019), the way the entrepreneurial discovery process is organized and implemented will show the ability of regions to diversify successfully. Besides capabilities, this depends on institutional and political leaders (Battilana et al. 2009) that trigger new initiatives, promote collective action, and induce institutional change (Garud et al. 2002). Institutional agents are considered crucial for the emergence of new activities in a region, because they collectively mobilize resources, build legitimacy, create new institutions or shape existing institutions (Sotarauta and Pulkkinen 2011). Such agents of change operate in institutional contexts that vary widely across regions, such as local governance cultures (Kroll 2015) and quality of government (Rodríguez-Pose and Di Cataldo 2015). A final note is that policy accounts in EEG also need to accommodate the decline and exits of activities in a region. As EEG is concerned with processes of creative destruction in regions, its prime focus should not be restricted to the creation and upgrading of local activities only (and which ones to prioritize in S3 policy). When regions are confronted with activities in decline (possibly as the result of the same creative process), studies have shown that a local supply of activities skill-related to the declining activities functions as a good shock-absorber, because such supply enhances regional labor matching (Neffke and Henning 2013; Diodato and Weterings 2015; Holm et al. 2017). It also prevents the 336

Evolutionary economic geography and policy

destruction of local human capital and the outflow of high-skilled people to elsewhere. However, in case of declining activities that are unrelated to existing local activities, this implies strong policy intervention is needed that focuses on re-education and reintegration of redundant workers in the local labor market.

References Aghion, P. and Bircan, C. (2017) The middle-income trap from a Schumpeterian perspective, EBRD Working Paper no. 205, 10.2139/ssrn.3119146 Alshamsi, A., Pinheiro, F. L. and Hidalgo, C. A. (2018) Optimal diversification strategies in the networks of related products and of related research areas, Nature Communications 9 (1), 1–7. Antonelli, C., Crespi, F. and Quatraro, F. (2022) Knowledge complexity and the mechanisms of knowledge generation and exploitation: The European evidence, Research Policy, 51 (8), 104081, doi: 10.1016/j.respol.2020.104081. Aranguren, M., Magro, E., Navarro, M. and Wilson J. (2019) Governance of the territorial entrepreneurial development process: looking under the bonnet of RIS3, Regional Studies 53, 451–461. Asheim, B. T., A. Isaksen and M. Trippl (2019) Advanced introduction to regional innovation systems, Cheltenham: Edward Elgar. Balland, P.-A., R. Boschma, J. Crespo and D. Rigby (2019) Smart specialization policy in the EU: Relatedness, knowledge complexity and regional diversification, Regional Studies 53 (9), 1252–1268. Balland, P. A. and Rigby, D. (2017). The geography of complex knowledge. Economic Geography 93(1), 1–23. Battilana, J., Leca, B. and Boxenbaum, E. (2009) How actors change institutions: Towards a theory of institutional entrepreneurship, The Academy of Management Annals, 3, 65–107. Barca, F. (2009) An Agenda for a Reformed Cohesion Policy: A Place-based Approach to Meeting European Union Challenges and Expectations, Report, Brussels: European Commission. Boschma, R. A. (2017) Relatedness as driver behind regional diversification: A research agenda, Regional Studies 51 (3), 351–364. Boschma, R. (2022) Designing smart specialization policy: relatedness, unrelatedness, or what? in: M. Anderson, C. Karlsson and S. Wixe (eds.), Handbook of Spatial Diversity and Business Economics, Oxford: Oxford University Press, forthcoming. Boschma, R., Coenen, L., Frenken, K. and Truffer, B. (2017) Towards a theory of regional diversification: Combining insights from evolutionary economic geography and transition studies, Regional Studies 51 (1), 31–45. Boschma, R. A. and Frenken, K. (2006) Why is economic geography not an evolutionary science? Towards an evolutionary economic geography, Journal of Economic Geography 6 (3), 273–302. Boschma, R. and Frenken, K. (2018) Evolutionary Economic Geography, in: G. Clarke, M. Feldman, M. Gertler and D. Wojcik (eds.), New Oxford Handbook of Economic Geography, Chapter 11, Oxford: Oxford University Press, pp. 213–229. Boschma, R. A. and Lambooy, J. G. (1999) Evolutionary economics and economic geography, Journal of Evolutionary Economics 9, 411–429. Boschma, R. and Martin, R. (eds) (2010) The Handbook of Evolutionary Economic Geography, Cheltenham: Edward Elgar. Boschma, R., Miguelez, E., Moreno, R. and Ocampo-Corrales, D. B. (2023) The role of relatedness and unrelatedness for the geography of technological breakthroughs in Europe, Economic Geography, 99 (2), 117-139 Breschi, S., Lissoni, F. and Malerba, F. (2003) Knowledge-relatedness in firm technological diversification, Research Policy 32 (1), 69–87. Castaldi, C., Frenken, K. and Los, B. (2015) Related variety, unrelated variety and technological breakthroughs: An analysis of US state-level patenting, Regional Studies 49 (5), 767–781. D’Adda, D., Guzzini, E., Iacobucci, D. and Palloni, R. (2019) Is Smart Specialisation Strategy coherent with regional innovative capabilities?, Regional Studies 53 (7), 1004–1016. Davies, B. and Maré, D. C. (2021) Relatedness, complexity and local growth, Regional Studies 55 (3), 479–494.

337

Ron Boschma Deegan, J., Broekel, T. and Fitjar, R. D. (2021) Searching through the Haystack. The relatedness and complexity of priorities in smart specialization strategies, Economic Geography 97 (5), 497–520. De Vaan, M., Frenken, K. and Boschma, R. (2019) The downside of social capital in new industry creation, Economic Geography 95 (4), 315–340. Di Cataldo, M., Monastiriotis, V., and Rodriguez‐Pose, A. (2022) How “Smart” Are Smart Specialization Strategies? Journal of Common Market Studies 60 (5), 1272-1298, 10.1111/jcms.13156. Diodato D. and Weterings A. B. R. (2015) The resilience of regional labour markets to economic shocks: Exploring the role of interactions among firms and workers, Journal of Economic Geography 15 (4), 723–742. Doner, R. F. and Schneider, B. R. (2016) The Middle-Income Trap: More Politics than Economics, World Politics 68 (4), 608–644. Edler, J., and Fagerberg, J. (2017) Innovation policy: What, why, and how, Oxford Review of Economic Policy 33 (1), 2–23. Evenhuis, E. (2016) The political economy of adaptation and resilience in old industrial regions: a comparative study of South Saarland and Teesside, PhD thesis, Newcastle University, Newcastle. Fleming, L. and Sorenson, O. (2001) Technology as a complex adaptive system: Evidence from patent data. Research Policy 30, 1019–1039. Florida, R. L. and Kenney, M. (1988) Venture capital, high technology and regional development, Regional Studies 22, 33–48 Foray, D. (2015) Smart Specialisation: Opportunities and Challenges for Regional Innovation Policy, London: Routledge. Foray, D. (2019) In response to ‘Six critical questions about smart specialisation’, European Planning Studies 27 (10), 2066–2078. Foray, D., David, P. A., and Hall, B. H. (2009) Smart specialization. The concept (Knowledge Economists Policy Brief No. 9, June), Brussels: European Commission. Foray, D., David, P. A., and Hall, B. H. (2011) Smart Specialization. From Academic Idea to Political Instrument, the Surprising Career of a Concept and the Difficulties Involved in its Implementation (MTEI Working Paper, November), Lausanne: MTEI. Foray, D., Goddard, J., Beldarrain, X. G., Landabaso, M., McCann, P., Morgan, K., Nauwelaers, C., Ortega-Argilés, R. and Mulatero, F. (2012) Guide to Research and Innovation Strategies for Smart Specialisations (RIS 3), Brussels. Fratesi U., Gianelle, C. and Guzzo, F. (2021) Assessing Smart Specialisation: Policy Implementation Measures, EUR 30758 EN, Publications Office of the European Union, Luxembourg. Frenken, K. (2017) Complexity-Theoretic Perspective on Innovation Policy, Complexity, Governance & Networks, 35–47, DOI: 10.20377/cgn-41 Frenken, K., Van Oort, F. G. and Verburg, T. (2007) Related variety, unrelated variety and regional economic growth, Regional Studies 41, 685–697. Fritsch, M. and Wyrwich, M. (2014) The long persistence of regional levels of entrepreneurship: Germany 1925 to 2005, Regional Studies 48, 955–973. Garud, R., Jain, S. and Kumaraswamy, A. (2002) Institutional entrepreneurship in the sponsorship of common technological standards: The case of Sun Microsystems and Java, Academy of Management Journal 45 (1), 196–214. Grabher, G. (1993) The weakness of strong ties: the lock-in of regional development in the Ruhr area, in Grabher G. (ed.), The Embedded Firm, London: Routledge, pp. 255–277. Grillitsch, M., Asheim, B. and Trippl, M. (2018) Unrelated knowledge combinations: The unexplored potential for regional industrial path development, Cambridge Journal of Regions, Economy and Society, 11 (2), 257–274. Hartmann, D., Bezerra, M., Lodolo, B. and Pinheiro, F. L. (2020) International trade, development traps, and the core-periphery structure of income inequality, EconomiA 21, 255–278. Hassink, R. (2005) How to unlock regional economies from path dependency? From learning region to learning cluster, European Planning Studies 13, 521–535. Hausmann, R. and Rodrik, D. (2003) Economic development as self-discovery, Journal of Development Economics 72 (2), 603–633. He, C. and Zhu, S. (2019) Evolutionary Economic Geography in China, Berlin: Springer. Henning, M. (2019) Time should tell (more). Evolutionary economic geography and the challenge of history. Regional Studies 53 (4), 602–613.

338

Evolutionary economic geography and policy Hidalgo, C., Balland, P. A., Boschma, R., Delgado, M., Feldman, M., Frenken, K., Glaeser, E., He, C., Kogler, D., Morrison, A., Neffke, F., Rigby, D., Stern, S., Zheng, S., and Zhu, S. (2018) The principle of relatedness, Springer Proceedings in Complexity, 451–457. Hidalgo, C. and Hausmann, R. (2009) The building blocks of economic complexity, Proceedings of the National Academy of Sciences 106, 10570–10575. Holm, J. R., Østergaard, C. R. and Olesen, T. R. (2017) Destruction and reallocation of skills following large company closures, Journal of Regional Science 57 (2), 245–265. Iacobucci, D. and Guzzini, E. (2016). Relatedness and connectivity in technological domains: Missing links in S3 design and implementation. European Planning Studies, 24 (8), 1511–1526. Janssen, M. and Frenken, K. (2019) Cross-specialisation policy: Rationales and options for linking unrelated industries, Cambridge Journal of Regions, Economy and Society 12, 195–212. Klepper, S. (2010) The origin and growth of industry clusters: The making of Silicon Valley and Detroit, Journal of Urban Economics 67, 15–32. Kroll, H. (2015) Efforts to implement smart specialization in practice – Leading unlike horses to the water, European Planning Studies 23 (10), 2079–2098. Kuhlmann, S. and Rip, A. (2018), Next generation innovation policy and grand challenges, Science and Public Policy 45 (4), 448–454. Lee, K. (2013) Schumpeterian Analysis of Economic Catch-up: Knowledge, Path-Creation, and the Middle-Income Trap, Cambridge: Cambridge University Press. MacKinnon, D., Cumbers, A., Pike, A., Birch, K. and McMaster, R. (2009) Evolution in Economic Geography: Institutions, Political Economy, and Adaptation, Economic Geography 85 (2), 129–150. Marrocu, E., Paci, R., Rigby, D., and Usai, S. (2022) Evaluating the implementation of Smart Specialisation policy, Regional Studies 57 (3), 1-17. Martin, R., and Sunley, P. (2006) Path dependence and regional economic evolution, Journal of Economic Geography 6 (4), 395–437. Maskell, P. and Malmberg, A. (1999) The competitiveness of firms and regions: ‘Ubiquitification’ and the importance of localized learning, European Urban and Regional Studies 6 (1), 9–25. Matti, C., Consoli, D. and Uyarra, E. (2017) Multi-level policy mixes and industry emergence: The case of wind energy in Spain, Environment and Planning C: Pol. Space 35 (4), 661–683. McCann, P. and Ortega-Argilés, R. (2015) Smart specialization, regional growth and applications to European Union Cohesion Policy, Regional Studies 49 (8), 1291–1302. McCann, P. and Ortega-Argilés, R. (2016) The early experience of Smart Specialisation implementation in EU Cohesion Policy, European Planning Studies 24 (8), 1407–1427. Mazzucato, M. (2013) The Entrepreneurial State: Debunking Public vs. Private Sector Myths, London/ New York: Anthern Press. Mazzucato, M. (2018) Mission-oriented innovation policies: Challenges and opportunities, Industrial and Corporate Change 27 (5), 803–815. Morgan, K. (2015) Smart specialisation: Opportunities and challenges for regional innovation policy, Regional Studies, 49 (3), 480–482. Muneepeerakul, R., Lobo, J., Shutters, S. T., Gomez-Lievano, A. and Qubbaj, M. R. (2013). Urban economies and occupation space: Can they get ‘there’ from ‘here’? PLoS ONE 8 (9), e73676. doi: 10.1371/journal.pone.0073676 Neffke, F. and Henning, M. (2013) Skill relatedness and firm diversification, Strategic Management Journal 34 (3), 297–316. Neffke, F., Henning, M. and Boschma, R. (2011) How do regions diversify over time? Industry relatedness and the development of new growth paths in regions, Economic Geography 87 (3), 237–265. Neffke, F., Hartog, M., Boschma, R. and Henning, M. (2018) Agents of structural change. The role of firms and entrepreneurs in regional diversification, Economic Geography 94, 23–48. Nelson, R. R., Dosi, G., Helfat, C. E., Pyka, A., Saviotti, P. P., Lee, K., Dopfer, K., Malerba, F. and Winter, S. G. (eds.) (2018) Modern Evolutionary Economics. An Overview, Cambridge: Cambridge University Press. Pagliacci, F., Pavone, P., Russo, M. and Giorgi, A. (2020) Regional structural heterogeneity: Evidence and policy implications for RIS3 in macro-regional strategies, Regional Studies 54, 765–775. Petralia, S., Balland, P-A. and Morrison, A. (2017) Climbing the ladder of technological development. Research Policy 46, 956–969.

339

Ron Boschma Pinheiro, F. L., Hartmann, D., Boschma, R. and Hidalgo, C. A. (2022a) The time and frequency of unrelated diversification, Research Policy 51 (8), 104323, doi: 10.1016/j.respol.2021.104323 Pinheiro, F. L., Balland, P. A., Boschma, R. and Hartmann, D. (2022b) The dark side of the geography of innovation: Relatedness, complexity, and regional inequality in Europe, Regional Studies, doi: 10.1080/00343404.2022.2106362 Rigby, D. (2015) Technological relatedness and knowledge space: Entry and exit of US cities from patent classes, Regional Studies 49 (11), 1922–1937. Rigby, D. L., Roesler, C., Kogler, D., Boschma, R. and Balland, P. A. (2022) Do EU regions benefit from Smart Specialisation principles?, Regional Studies 56 (12), 2058-2073, DOI: 10.1080/00343404. 2022.2032628. Rodríguez-Pose, A. and Di Cataldo, M. (2015) Quality of government and innovative performance in the regions of Europe, Journal of Economic Geography 15 (4), 673–706. Rodríguez-Pose, A. and Wilkie, C. (2015) Institutions and the entrepreneurial discovery process for smart specialization (Papers in Evolutionary Economic Geography No. 15.23). Utrecht: Utrecht University. Rodrik, D. (2004) Industrial Policy for the Twenty-First Century, Cambridge MA: Harvard University, Kennedy School of Management Working paper n. rwp04-047. Schot, J. and Steinmueller, W. E. (2018) Three frames for innovation policy: R&D, systems of innovation and transformative change, Research Policy 47 (9), 1554–1567. Sotarauta, M. and Pulkkinen, R. (2011) Institutional entrepreneurship for knowledge regions: In search of a fresh set of questions for regional innovation studies, Environment and Planning C 29, 96–112. Tanner, A. N. (2016) The emergence of new technology-based industries: The case of fuel cells and its technological relatedness to regional knowledge bases, Journal of Economic Geography 16 (3), 611–635. Teece, D. J., Rumelt, R., Dosi, G., and Winter, S. (1994). Understanding corporate coherence, Journal of Economic Behavior and Organization 23, 1–30. Teece, D. J., Pisano, G., and Shuen, A. (1997). Dynamic capabilities and strategic management, Strategic Management Journal 18 (7), 509–533. Tödtling, F. and Trippl, M. (2005) One size fits all? Towards a differentiated regional innovation policy approach, Research Policy 34 (8), 1203–1219. Trippl, M., Zukauskaite, E. and Healy, A. (2020) Shaping smart specialization: The role of placespecific factors in advanced, intermediate and less-developed European regions, Regional Studies 54 (10), 1328–1340. Van den Berge, M., Weterings, A. and Alkemade, F. (2020) Do existing regional specialisations stimulate or hinder diversification into cleantech? Environmental Innovation and Societal Transitions 35, 185–201. Xiao, J., Boschma, R. A. and Andersson, M. (2018) Industrial diversification in Europe: The differentiated role of relatedness, Economic Geography 94 (5), 514–549. Zhu, S., He, C. and Zhou, Y. (2017) How to jump further and catch up? Path-breaking in an uneven industry space, Journal of Economic Geography 17 (3), 521–545.

340

26 GLOBAL KNOWLEDGE EMBEDDEDNESS Holger Graf and Martin Kalthaus

26.1

Introduction

The process of knowledge generation, invention and innovation is cumulative and inter­ active (Ahuja, 2000; Breschi & Lissoni, 2004; Dosi, 1988; Powell et al., 1996). New ideas are developed based on existing knowledge, and the division of labor leads to specialists in science and research who need to collaborate in increasingly large teams (Fortunato et al., 2018; Jones, 2009; Wuchty et al., 2007). This process of knowledge collaboration and ex­ change has become increasingly globalized, with a worldwide expansion of mass higher education, growth in the number of international student mobility (Shields, 2013), greater migration of scientists and engineers (Freeman, 2010) and more international co-authorship (Glänzel & Schubert, 2005) and co-patenting (Picci, 2010). All these forms of interaction constitute a global network of knowledge generation and diffusion (Adams, 2013, 2012; Keller, 2004). Embeddedness therein is a necessary asset for any individual, organization and country to be successful in idea generation, invention and innovation. In this chapter, we review the literature on two of the most widely studied channels of international knowledge diffusion in the field of science: research collaboration and scientist mobility. We thereby focus on the motives to collaborate or move internationally, the ef­ fects of such actions and how they lead to aggregate outcomes, in particular, global network structures. Based on this, we discuss how embeddedness in such networks influences knowledge generation and performance. Embeddedness in the global knowledge network is captured by the position of actors, e.g. individuals, organizations or countries, in the network of knowledge related interac­ tions between these actors. In general, higher embeddedness facilitates access to knowledge with positive performance effects (Uzzi, 1996, 1997; Wanzenböck et al., 2014, 2015). The global network is constituted by the knowledge exchange of these actors across territorial borders. Conceptually, such interactions cover a variety of dimensions, such as formality, temporality, intensity, frequency or purpose (Georghiou, 1998), but to observe and measure interaction in all these dimensions is a challenge (Katz & Martin, 1997; Laudel, 2002; Melin & Persson, 1996; Teichler, 2015). Interactions that are most easily observed include joint publications or mobility of individuals between places and/or organizations. Due to their

DOI: 10.4324/9780429398971-29

341

Holger Graf and Martin Kalthaus

intensity of face-to-face exchange, these interaction channels are also considered particu­ larly effective. Other forms of interaction at meetings or conferences, sharing of data or methods or other means of exchange are harder to track, potentially more frequent but less important for knowledge exchange (e.g. Cronin et al., 2003, 2004; Laudel, 2002). In the following, we summarize the empirical and conceptual development of the global knowledge network by co-authorship and mobility. In Section 26.3, we review the literature on drivers of collaboration and mobility on the level of the individual researcher. Aggregate patterns, determinants and consequences of global knowledge embeddedness are discussed in Section 26.4. We summarize our findings in Section 26.5 by deriving a set of stylized facts and presenting our ideas on fruitful avenues for future research.

26.2

Knowledge diffusion through collaboration and mobility

Over the past decades, we observe a continuous increase in interaction and collaboration across all fields of science and research (Wuchty et al., 2007). One of the reasons for this development is an increasing specialization and division of labor because of the cumulative and dispersed nature of knowledge (Jones, 2009). Vast empirical evidence indicates that collaborative research leads to more valuable output than individual research (e.g. Adams, 2013; Adams et al., 2005; Jones, 2021; Wuchty et al., 2007). Research teams and co-authors do not just add their individual expertize to generate joint output, but they also exchange information and learn from each other (Breschi & Lissoni, 2004). The increasing and sys­ tematic collaboration and mobility of scientists can be seen as a result of increasing pro­ fessionalization of science (Beaver & Rosen, 1978; Beaver & Rosen, 1979). Scientific knowledge grows exponentially and the problems researchers address become more com­ plex, requiring higher degrees of interaction and exchange. This process has been referred to as the transition from ‘little science’ to ‘big science’ (Solla Price, 1963). Larivière et al. (2015) show for publication data from 1900 onwards that co-authorship is increasing substantially and is the norm nowadays. Along with this development, international research collabo­ ration increases continuously as well. The development and the structures of interaction among researchers through collabo­ ration and mobility have been subject to several approaches to conceptualize the phe­ nomenon. From a science of science perspective, the increasing professionalization in science and the transition from little science to big science can be organized in an ‘invisible college’ (Price & Beaver, 1966; Solla Price, 1963). The invisible college governs the scientific community and facilitates formal and informal knowledge exchange globally and brings together the members of the community (Crane, 1972). This self-governing process of the researcher community has more recently been referred to as the ‘fourth age of research’ (Adams, 2013) in which international collaboration between research groups form to approach scientific problems. Prominent examples of this tendency are large research facilities, such as CERN, where multiple international teams conduct research together (Carrazza et al., 2016). In these invisible colleges, preferential attachment, i.e. the tendency to form linkages with already highly connected actors, is a key characteristic of interna­ tional collaboration (Wagner & Leydesdorff, 2005a). Within the economics of innovation, the concept of (national) innovation systems is extended and integrates international knowledge flows to account for the division of labor in science and the dispersed nature of knowledge (Bathelt et al., 2004; Niosi & Bellon, 1994). Due to the increasing importance of science in technological progress and economic growth, 342

Global knowledge embeddedness

countries need to participate in international knowledge flows (Ribeiro et al., 2018) and integration into these flows can be a starting point for developing countries to catch up. Differentiating between industries and technologies, Binz & Truffer (2017) propose that in multi-locational subsystems resources are acquired and generated and that the different groups of actors are coupled in a global innovation system to exchange knowledge and resources. The functional design of such a subsystem, can influence the embeddedness in the global system and allows countries to access global flows of knowledge via international collaboration (Graf & Kalthaus, 2018). From a policy perspective, governments influence international collaboration to foster the exchange of knowledge or to gain access to foreign knowledge. Already Bush (1945) emphasized that the U.S. government should foster the international flow of knowledge. Governments design programs supporting scientists to engage in international collabora­ tion or to move temporally abroad to establish contacts. The leading example is the European Research Area in which scientific collaboration and mobility are core compo­ nents to foster knowledge exchange and subsequent technological and economic progress (Defazio et al., 2009; European Union, 2016). This is realized especially via joint research programs (Balland et al., 2019) and mobility grants (Ackers, 2005). In the growth and development literature, an important stream of research is concerned with the geographical distribution and the direction of migration and mobility flows between countries. These flows have been found to be highly asymmetrical, not least because creative minds have always been attracted to particularly vibrant cities or regions that act as hubs in the global knowledge network (Doehne & Rost, 2021; Florida, 2005). The migration of skilled and educated workers from developing to developed countries was identified as detrimental for economic growth in the sender countries so the term “brain drain” was used to describe the phenomenon (e.g. Beine et al., 2008, 2001). Historical studies on large migration flows show a large impact on both, sender and receiver countries (e.g. Moser et al., 2014). The opposite, positive effects for the receiving countries are accordingly referred to as ‘brain gain’. However, there are also potential positive feedback effects of skilled migration on the sender countries, such as remittances, return migration or scientific and business networks. In addition, migration prospects might increase the incentives to invest in education and human capital formation in the home countries (Beine et al., 2008). Return migration and the role of diasporas have been studied, for example, by Agrawal et al. (2011) and Saxenian (2005), who argue that scientist migration establishes long-term connections between migrants and their home countries which might facilitate knowledge flows towards the sender countries. It was also found that emigration of students, academics and other skilled professionals is increasingly temporary (Gaillard & Gaillard, 1997) so the concept of ‘brain circulation’ has been suggested as a more complete model (Baruffaldi & Landoni, 2012; Jöns, 2009). China, for example, was able to benefit from international mobility and thereby strengthen its research and innovation system by sending scientists abroad and by being able to attract them back (Verginer & Riccaboni, 2020).

26.3 Drivers of international mobility and collaboration 26.3.1 Determinants of scientist mobility The reasons for migration and mobility are manifold and range from individual choices in search of better opportunities to forced migration due to religious persecution. Historical

343

Holger Graf and Martin Kalthaus

examples of larger waves of mobility of skilled workers or scientists illustrate their great impact on the global structure of knowledge diffusion and distribution. For example, Hornung (2014) finds long-term productivity effects of religiously persecuted French Huguenots who settled in Brandenburg-Prussia after the revocation of the Edict of Nantes in 1685. Moser et al. (2014) look at Jewish emigrants from Nazi Germany and identify a strong impact on invention in the United States. Besides such unique and often dramatic events, there is a continuous increase in mobility over the last decades, and this trend is particularly pronounced for scientists (Czaika & Orazbayev, 2018). According to their analysis of bibliometric data, the share of mobile scientists who moved internationally at any point in their career has been steadily increasing from about 6% in the 1970s to about 9% by the mid-2010s. So what are the drivers, the individual motivations and determinants of this voluntary mobility of scientists? A main driver seems to be the aspiration to work in a more professional or better equipped environment, thereby increasing their international professional network (Kato & Ando, 2017). Consequently, the presence of prestigious universities and research organi­ zations of high scientific quality is considered as an important factor for attracting scientists to particular cities, regions or countries (Bauder, 2015). Several studies show that prolific scientists are particularly mobile and choose their destination carefully by preferentially moving to global cities to continue their career (Azoulay et al., 2017; Verginer & Riccaboni, 2021). There is ample evidence for positive effects of the decision to become mobile on scientists’ productivity, scientific impact and occupational situation (Gibson & McKenzie, 2014; Netz et al., 2020; Verginer & Riccaboni, 2021). Scellato et al. (2015) are particularly interested in the relationship between scientist mobility and their collaboration patterns. Their study is based on a very large survey among more than 15000 scientists from 16 countries in selected STEM fields. Their main finding is that migrants and returnees have larger international research networks than their native counterparts without any interna­ tional background. These mobile scientists often collaborate with researchers from their country of origin, whether they are in their home country or a third country. The mobilitycooperation nexus is also tackled by Kato & Ando (2017), who conduct a study based on metadata from papers published in Nature and Science between 1989 and 2009. They also find that researchers move to countries with better research environments. However, in line with Wang et al. (2019), they identify international mobility as a driver of collaboration but not as much in the other direction. Additionally, the benefits of mobility for knowledge diffusion are not limited to the academic sphere. Edler et al. (2011) show for a sample of German scientists that temporarily mobile researchers are involved in knowledge and technology transfer activities to industry in both home and host countries. They also find that the duration of research visits has a positive impact on the propensity to transfer knowledge and technology in both countries. Jöns (2007) studies the motives for temporary research stays in Germany and finds that besides the above-mentioned expectations regarding a better professional environment, scien­ tists are also attracted by existing professional contacts. Other motives that are frequently mentioned in this study are the search for new experiences and ideas and the time to do research and to publish academic work and contacts with foreign researchers. Another factor that de­ termines mobility patterns is the career stage. Bauder (2015) provides ample evidence that younger researchers, particularly those in their post doc phase, are more mobile than researchers at later stages of their career. Adding to such life-cycle arguments, Azoulay et al. (2017) find that having adolescent children is a strong barrier for mobility, in particular for mothers. 344

Global knowledge embeddedness

Besides factors relating to individual life and career planning, governing institutions and mobility support structures by the government or foundations increasingly promote inter­ national mobility, especially for young academics (Bauder, 2015; Jacob & Meek, 2013). One could ask if we still need to meet each other in person to effectively exchange knowledge and ideas. During the COVID-19 pandemic, everybody could experience the possibilities and also the limitations of online meetings, collaborations and conferences. So while technological developments in ICT certainly facilitate long-distance interaction, they are by no means a substitute for face-to-face interaction. Rather, technological develop­ ments seem to further reinforce the importance of face-to-face interaction (Czaika & Orazbayev, 2018). From the reviewed literature on the drivers of mobility, we see that the scientific network of collaborative ties is sometimes cause for but mostly the effect of sci­ entific mobility, showing how deeply intertwined these forms of interaction are in the global knowledge network.

26.3.2

Determinants of international collaboration

The increasing trend in the collaboration in knowledge production in teams (Wuchty et al., 2007), across organizations (Jones et al., 2008) and internationally (Adams, 2013; Wagner et al., 2017) is motivated by different factors and has several consequences. The general increasing trend in collaboration and team formation has many motivations and reasons, as summarized by, e.g. Katz & Martin (1997) and Beaver (2001). Since international collab­ oration can have higher transaction costs or higher barriers (Ou et al., 2012), researchers pursue international collaborations for specific reasons, or researchers have specific char­ acteristics which make them pursue collaboration across borders (Freeman et al., 2015). Frame & Carpenter (1979) conduct one of the first empirical assessments of international research collaboration and provide a list of determinants on the collaboration intensity and partner choice for international collaboration. They conclude that “[t]he size of a national research effort, and a number of non-science factors–including geographic locale, and lin­ guistic, cultural, and political factors …” (p. 495) are relevant for the intensity and direction of international collaboration. This list of determinants has been extended with, e.g. the decline in travel cost, improvements in communication technologies, the rise of English as the common language in science, governmental programs, division of labor and specialization, joint research infrastructures, cultural traditions and norms and values (e.g. Luukkonen et al., 1992; Wagner & Leydesdorff, 2005a; Waltman et al., 2011). However, the relevance of these determinants changes over time. For example, geographical proximity and territorial borders have lost relevance over time (Hoekman et al., 2010). Wagner et al. (2015) show that the share of international collaborations and the geographical distance between co-authors increases over time, and Waltman et al. (2011) calculate that the average collaboration distance per publication has increased fivefold from 1980 to 2009. Similarly, Catalini et al. (2020) show that a decline in air-travel costs increases collaboration. Furthermore, characteristics or motivations of individual scientists can explain inter­ national collaboration. For example, Bozeman & Corley (2004) survey U.S. scientists and show that many collaborations are local or national but that researchers with large grants show a more ‘cosmopolitan’ collaboration pattern with contacts to industry and interna­ tional research partners. Based on another U.S. survey, Melkers & Kiopa (2010) analyze the motivations and determinants for international collaboration. They hypothesize that resources and social capital facilitate international collaboration. They find positive 345

Holger Graf and Martin Kalthaus

relationships between a researcher’s international experiences and individual characteristics, such as foreign nationality, and international collaboration, highlighting the role of social capital. The researcher’s access to resources, as well as the reputation of the researcher’s institute, also correspond with international collaboration. They further show that such international collaboration allows researchers to access specific resources. Similar results are provided by Jeong et al. (2014) for internationally co-authored papers on Korean research projects. In a similar vein, Jöns (2009, 2007) emphasizes the importance of previous per­ sonal interaction, publication and (temporal) mobility for international collaboration. However, gender differences and specific barriers for women exist (Fox et al., 2017). Extrinsic incentives, such as R&D subsidies and funding, have also been shown to increase collaboration (Adams et al., 2005; Ebadi & Schiffauerova, 2013). The scientific benefits from international collaboration are an additional strong moti­ vator for researchers to engage in such collaborations. A broadly used measure for the success of scientific articles are received citations. By distinguishing between papers with multiple authors from multiple countries, one can assess the scientific benefits of interna­ tional collaborations. Narin et al. (1991) were among the first who analyzed the influence of international collaboration on scientific publications. They show that publications with authors from different countries have higher citation rates than publications from authors from the same country. Thereby, there are no differences between international publications from EU-EU and EU-non-EU countries. Glänzel (2001) finds a similar pattern on the aggregate level, but heterogeneous development between disciplines and country pairs, e.g. some country pairs of international collaboration show no observable increase in citations. Besides the increase in citations for international co-authored publications, Persson et al. (2004) show that such papers have about two references more than non-international, coauthored papers, indicating a higher rate of knowledge recombination. Larivière et al. (2015) also show, controlling for self-citation, that papers with international co-authorship receive more citations. However, several contributions challenge the higher impact of international co-authored publications. Leeuwen (2009) find on average a lower impact of internationally co-authored papers than single-authored papers but substantial heteroge­ neity on the individual country level. Similarly, Freeman et al. (2015) find heterogeneous effects across disciplines. In a different way, Wagner et al. (2019) analyze the novelty of papers written by international teams and show that such papers are less novel and contain more conventional knowledge combinations. They also argue that higher citations for international publications go back to an audience effect, since an increase in authors from different countries increases the potential citing community. In a different approach, Leung (2013) shows for Chinese researchers in nanomedicine that international collaborations can, besides an increase in resources and reputation, also provide opportunities for learning-bydoing which allow Chinese researchers to build up absorptive capacity. This suggests that the effects of international collaboration unfold over time.

26.4 26.4.1

The global network of knowledge embeddedness From individual interactions to a global structure

The increasing mobility of individuals and collaboration across national borders can be aggregated into a global network of knowledge exchange (Glänzel & Schubert, 2005). This knowledge network can be considered a hierarchical construct with three interdependent

346

Global knowledge embeddedness International collaboration between countries Macro

Collaboration between organizations Meso

Co-authorship at the researcher level Micro

Figure 26.1

Example of multimodal structures as in Graf & Kalthaus (2018).

layers based on the conceptualization of a micro-meso-macro structure (Dopfer et al., 2004; Graf & Kalthaus, 2018). On the micro level, the individual researcher makes a relocation or collaboration decision and selects his or her destination or partner. These individual re­ searchers can thereby establish links across national borders. Individual researchers can be aggregated based on their organizations to a meso level, where the links of individuals establish connections between organizations. The subsequent macro level is the aggregation of organizations in one country, and countries are connected via international collaboration or mobility of researchers belonging to the organizations in a country. Figure 26.1 shows a graphical representation of the different levels and aggregations of the knowledge network. For each level of aggregation of the knowledge network, structural properties, such as small world properties, or processes of network formation, such as preferential attachment or homophily, can be analyzed (Barabási et al., 2002; Newman, 2001; Wagner & Leydesdorff, 2005a). Besides the network structure and its dynamics, the position of individual nodes, the individual researcher, the organization or the country and its effects are of interest. Freeman (1979) argues that a central position relates to importance or power in a network, since it allows control of information flows between otherwise unrelated actors. Furthermore, some positions within the knowledge network might give an advantage for accessing novel or diverse knowledge. Understanding the changes in network position and the influence of network positions on performance is highly relevant to access and utilize knowledge flows in the network. Empirical evidence for innovation networks shows that both direct and indirect connections matter for research and innovation performance (for reviews, see Cantner & Graf, 2011; Hidalgo, 2016; Ozman, 2009; Phelps et al., 2012). Only recently has the interre­ latedness between the different levels of knowledge networks within multimodal structures been tackled in empirical studies (e.g. Graf & Kalthaus, 2018; Guan et al., 2015).

26.4.2

Patterns and dynamics of the global knowledge network

The pattern and dynamics of country networks have been studied extensively. In such net­ works, countries are nodes and connected via co-authorship between researchers from dif­ ferent countries or the mobility of researchers across countries. Schubert & Braun (1990) were among the first who use co-authorship information to construct a network of countries. They show for the period 1981–1985 a core of Western developed countries and some clusters of countries in the periphery. They argue that besides geographical location, historical or 347

Holger Graf and Martin Kalthaus

political reasons generate such a pattern. In a comparison of network structures in 1990 and 2000, Wagner & Leydesdorff (2005b) show that the network grew in size and connectedness and that the core of scientifically advanced countries in the network expanded. However, they also observe that the network becomes more decentralized and regional clusters emerge. Similar findings are obtained by Czaika & Orazbayev (2018) for the global scientific mobility network. They use bibliometric data for the period 1970 to 2014 to calculate annual mobility events for individual authors and aggregate them to the country level. Over time, an increasing number of countries integrates into the global network, and they observe shifts from the periphery to the core for several countries, with average mobility distances increasing and the center of gravity moving eastwards. At a more fine-grained level, Verginer & Riccaboni (2020) look at the global city network based on mobility data for the period 1990 to 2009. They identify a highly connected, relatively stable core of cities, with most of the hubs being located in the US but pronounced catching up by Beijing. Gui et al. (2019) conduct a detailed analysis of the international collaboration network for the period 2000 until 2015 and show also an increase in the number of countries, with nearly all countries being involved in international collaboration. Most frequent are col­ laborations between the USA and European countries, Japan and Australia. European countries also collaborate frequently among each other. In general, the number of inter­ connections increases nearly threefold in this time period. Consequentially, the mean degree and the density in the network increases over time, and nearly 50% of the possible country connections are established in 2015. The network is highly centralized, but over time, some decentralization takes place, and the scientifically advanced countries lose some of their centrality in the network. Gradually, the general structure changes from a bi-polar core network between Europe and North America to a tri-polar network in which the AsiaPacific region becomes more prominent. Thereby, collaboration between the USA and China is most prominent. Gui et al. (2019) furthermore analyze the core-periphery structure in the global network and categorize countries into four groups. They observe that over time, countries are moving upwards in the hierarchy and become closer to the core. Most countries in the periphery are lower-income or small countries. Additionally, the authors apply dominant flow analysis and show that the USA is the dominant coordinating actor in the network, only a few sub-dominant actors exist and all other countries are directly affiliated to the USA or the sub-dominated countries. Even though they observe dynamics with several countries changing their relative positions, the dominating role of the USA remains unchallenged. While these findings show the structure and dynamics for the overall scientific system, field-specific patterns can substantially differ. For example, Gazni et al. (2012) show that in the period 2000–2009, 44% of all space science papers included international collaborations, while in social sciences the rate was only 6%. Wagner et al. (2017) show for six disciplines for the period 1990 to 2013 a substantial growth in field-specific network size and con­ nectivity, however, with high differences in their levels. Mathematics is the smallest and least connected network, with even a decrease in connectedness at the end of the obser­ vation period. In contrast, astrophysics has the largest and most densely connected inter­ national network, in which countries have on average eight times more connections to other countries than in mathematics. They conclude that over time, the fields converge towards the global development of a highly connected science system. Similar results have been shown by Coccia & Wang (2016) for a period of 40 years. They find that there is an increase and convergence between applied and basic sciences in terms of internationalization but 348

Global knowledge embeddedness

that the ranking in international collaboration intensity by field does not change over time. Overall, these findings indicate that despite differences in the intensity of interaction, globalization trends are ubiquitous across science fields. With respect to the partners involved in international collaborations, Pohl (2021) analyzes collaboration between aca­ demia and corporations. He shows that in relative terms, the share of such collaborations does not increase and accounts for about 2.5% of all publications.

26.4.3 Determinants of country embeddedness Several arguments have been put forward to explain why countries form linkages in research. Zitt et al. (2000) show for five core countries that their likelihood to form an international collaboration is influenced by political and cultural, linguistic, economic and geographical factors. Gui et al. (2019) empirically test ten different factors on the likelihood that a connection is established between two countries. They find no influence of geographic distance, but having English as an official language and a colonial relationship in the past increase the likelihood to establish a connection. Furthermore, international student mobility is highly relevant, as is the difference in the number of publications and the dif­ ference in the research expenditure per country. Lastly, scientific, social and economic proximity are relevant to establish a collaboration. Similar results are reported by Hou et al. (2021) who subsume factors under scientific, economic, geopolitical and cultural factors. Using a relative measure of closeness in collaboration, they show for six disciplines that the more similar countries are in these factors, the closer they are collaborating, while the larger are the differences in scientific and economic capabilities as well as geographic distance, the lower is the closeness. They also report detailed, heterogeneous influences of different languages, religions and geopolitical factors across disciplines on collaboration closeness. In a different methodological approach, Plotnikova & Rake (2014) consider pharma­ ceutical research and test in a gravity model for the influence of geographical, cognitive, institutional, social and cultural proximity on formation of international collaboration. They show that geographical proximity has a negative effect but social proximity a positive one. Wagner & Leydesdorff (2005a) provide a different perspective and argue that inter­ national collaboration is governed by preferential attachment and that the science system is self-organizing. Their results point towards scale-free network properties, consistent with Barabási et al. (2002) or Ribeiro et al. (2018) for collaboration in scientific fields. Furthermore, universities use internationalization as a strategic element to increase em­ beddeddness and facilitate knowledge flows. Youtie et al. (2017) and Kolesnikov et al. (2019) show how universities strategically set up research or teaching locations in foreign countries to institutionalize international research collaborations. Similar factors are also relevant for the international mobility of scientists. Appelt et al. (2015) use changes in the affiliation of researchers’ publications over the period 1996–2011 to track mobility and explain aggregate mobility between origin and destination countries by economic, cultural and scientific factors. They show that scientific and economic distance between two countries reduces mobility, indicating that a convergence of countries in these dimensions increases mobility between the countries. Furthermore, they show that geo­ graphic distance and visa restrictions have a negative influence on mobility, while resources dedicated to R&D increase mobility. Besides these factors, research collaboration is a major factor for scientists’ mobility, indicating the co-determination between collaboration and mobility. They conclude that mobility forms a complex international network and that there 349

Holger Graf and Martin Kalthaus

is ample evidence for brain circulation. However, Arrieta et al. (2017) use the case of the European integration to show that access to the Western European labor market for Eastern European scientists increases their mobility westwards and subsequently reduces international collaboration rates for Eastern European countries. From a network position perspective, Cantner & Rake (2014) analyze endogenous net­ work dynamics that establish or reinforce international collaboration between countries in pharmaceutical research. They show that tie formation and tie break-up are related to countries’ relative connectedness, i.e. countries that have a larger difference in their relative positions in the network are more likely to connect. Furthermore, they demonstrate that the number of past collaborations of a country increases the likelihood that a connection is established, indicating that collaborative activity is self-reinforcing and cumulative. Country similarity in terms of scientific output increases collaboration, but economic differences between countries are not relevant. Sharing the same language shows ambiguous results in their analysis. Lastly, they argue that multi-connectivity is relevant for tie formation. Empirically, they show that knowledge flows matter by demonstrating that countries which indirectly connect two countries increase their connections. For photovoltaics research, Graf & Kalthaus (2018) show for the period 1980–2015 that the embeddeddness of coun­ tries in the global network is influenced by the structure and functionality of the underlying national research system and by policy intervention. The embeddeddness into the global knowledge network is measured by countries’ degree, flow betweenness and coreness. In that paper, we showed that cohesion and connectedness of the national research system positively affect international embeddedness, whereas centralized national research systems are detrimental to international embeddedness. This indicates that a diffusion-oriented research system allows better access to international knowledge flows. We also show that demand-inducing innovation policy can increase international embeddedness. Related research shows that on the national level, research universities play a central role in facil­ itating access to and embeddedness in the global knowledge network, in particular for lowand middle-income countries (Altbach, 2013). By taking the example of South Korea, Ahn et al. (2019) show that universities are key for accessing global knowledge flows and that they can bridge knowledge flows to other domestic universities, which are subsequently able to increase their performance.

26.4.4

Embeddedness and performance

A country’s level of international collaboration and its position in the global knowledge network – be it by mobility or by collaboration linkages – improves access to knowledge flows with positive effects on scientific activity and performance. Cimini et al. (2016) as well as Wagner et al. (2017) show that there is a strong correlation between internationalization of science in a country and its success in terms of citations. In a methodologically different approach, Wagner et al. (2018) construct an openness index based on international col­ laborations and scientist mobility and show that the openness of countries correlates with its scientific impact. Such openness and connection to the global knowledge network is also documented for individual countries. Nguyen et al. (2017) show for Vietnam that research increased substantially after it opened up to the world in 1986, and after the USA revoked a trade embargo against Vietnam in 1995, most research output was conducted in interna­ tional collaborations, especially with scientifically advanced countries. They argue that such international collaborations were essential to develop research capacity. Over time, the 350

Global knowledge embeddedness

share of internationally co-authored papers declined as national capacity was built. Furthermore, papers with international collaboration receive more citations and are pub­ lished in higher ranked journals than domestic ones. In a detailed analysis of the country-level benefits from international scientist mobility, Verginer & Riccaboni (2020) show that not all countries benefit from international exchange. The authors decompose growth in scientific publications into growth effects attributed to stationary, domestically mobile and internationally mobile scientists. Interestingly, for most countries, international mobility has a higher weight than domestic mobility. Regarding the age structure, they find that for most countries, leaving scientists are younger than the incoming, typically more prolific ones. China, for example, benefits from interna­ tional mobility by sending scientists abroad and attracting them back. India, on the other hand, does not benefit from international mobility in terms of direct research output, neither do Israel or Japan. Apparently, the particular settings of the respective national innovation systems play not only a big role in determining the level of inter­ national embeddedness but also in how countries can benefit from it. When looking at the beneficiaries of international embeddedness from a micro perspective, there are some indications that knowledge diffusion is highly uneven. It is mostly direct collaborators who benefit from connections with mobile inventors (Zacchia, 2018), and in low income countries, such links are established by few individuals (Vanni et al., 2014). Findings on the city level also point to uneven distributions of knowledge flows within countries (Verginer & Riccaboni, 2021). There are also indications that such performance effects are not limited to academia, but the evidence here is still limited. Bathelt & Li (2020) analyze how Canadian firms tap into local knowledge in China and show that cross-border knowledge generation is possible, especially if the local knowledge in China is systematically integrated with corporate knowledge pools. Tsouri et al. (2021) take the case of offshore wind and show that em­ beddedness in international knowledge networks can also facilitate access to market resources in a Global Innovation System.

26.5

Conclusion

The world is moving closer together across all kinds of activities. Just as trade flows increase along ever more complex global value chains, sources of knowledge become more dispersed so that keeping up with scientific and technological developments requires integration within the global research and innovation system. Taking a network perspective, in this chapter, we reviewed the literature connected to the phenomenon of global knowledge embeddedness, with a particular focus on two of the main channels of international knowledge diffusion: mobility of and collaboration among scientists. A common trend in the literature is that increasingly sophisticated methods are employed to analyze large da­ tabases of scientific publications and patents to identify mobility and collaboration patterns over long time spans. From this review, we derive stylized facts on the structure and dynamics of the global knowledge network. 1 The global knowledge network changes continuously with an increasing participation (globalization) and connectivity. 2 The global knowledge network is characterized by a core-periphery structure, with a growing core and countries changing relative positions. 351

Holger Graf and Martin Kalthaus

3 Over time, there is a global shift from a bi-polar (Europe - North America) to a tri-polar world (Europe - North America - Asia). Consequently, the global network is becoming less centralized with larger spatial distances covered by mobility and collaboration linkages. 4 Networks across scientific fields differ substantially in their interaction intensity, but the observed globalization trends are ubiquitous. 5 The determinants for international collaboration change over time, with geographical and political factors (e.g. a common colonial history) becoming less relevant over time. International mobility is a main driver of international collaboration in science. 6 Network structures and innovation systems on different levels are interrelated. A func­ tioning national system influences not only embeddeddness but also how countries benefit from being embedded. 7 Being embedded in the global knowledge network and openness towards global knowl­ edge flows is positively correlated with scientific impact and performance. While our understanding of global patterns and dynamics of interaction has increased substantially, much is still to understand, especially on the determinants, consequences and effects of international knowledge embeddedness. First, we need a better under­ standing of why a collaboration or mobility decision is made and what the individual and contextual determinants are. In particular on the contextual level, the inter-relatedness between networks on different levels of aggregation is a fruitful avenue for future research. The organization of research systems as well as national strategies towards international knowledge exchange can significantly affect the access and flow of knowl­ edge. Understanding such relationships is also highly relevant from a policy perspective in the context of the brain-drain – brain circulation debate. Second, we need to learn more about the effects of embeddedness in the global knowledge network. While first results show that research performance increases due to international collaboration and mobility, direct effects on economic performance have not been analyzed yet. In this context, it would be of great importance to understand the mediating role of (national) preconditions, such as the absorptive capacity of nations, to benefit from growing international embeddedness. Improved access to and use of global knowledge flows for countries far from the scientific frontier could reduce knowledge inequality and help increase innovation and economic prosperity. A limitation of the literature is a lack of causal evidence on the determinants and effects of global knowledge embeddedness. Most of the insights presented in this chapter are based on correlations. Conditions that allow for a causal identification of knowledge flows, international embeddedness and the effects of such embeddedness are hard to identify. There are a few exceptions. Agrawal et al. (2016), for example, use the collapse of the Soviet Union and the sudden inflow of mathematical knowledge into the United States as a natural experiment. The COVID-19 pandemic can provide cases to analyze how global networks are formed or disintegrated, as, for example, by Wagner et al. (2021). As a final methodological remark, the realization of knowledge flows is only implicit in studies of the global network. The underlying assumption is that through relationships among individuals, knowledge is exchanged between countries so that subsequently, knowledge is diffused within the country and then exchanged again with other countries. A more explicit analysis of actual knowledge flows can support or invalidate the underlying assumptions, which would have significant implications for the relevance of global knowledge networks. 352

Global knowledge embeddedness

References Ackers, L. (2005). Promoting scientific mobility and balanced growth in the European Research Area. Innovation: The European Journal of Social Science Research, 18(3), 301–317. 10.1080/135116105 00186680 Adams, J. (2013). The fourth age of research. Nature, 497(7451), 557–560. 10.1038/497557a Adams, J. (2012). The rise of research networks. Nature, 490(7420), 335–336. 10.1038/490335a Adams, J. D., Black, G. C., Clemmons, J. R., & Stephan, P. E. (2005). Scientific teams and institu­ tional collaborations: Evidence from U.S. Universities, 1981–1999. Research Policy, 34(3), 259–285. 10.1016/j.respol.2005.01.014 Agrawal, A., Goldfarb, A., & Teodoridis, F. (2016). Understanding the changing structure of scientific inquiry. American Economic Journal: Applied Economics, 8(1), 100–128. 10.1257/app.20140135 Agrawal, A., Kapur, D., McHale, J., & Oettl, A. (2011). Brain drain or brain bank? The impact of skilled emigration on poor-country innovation. Journal of Urban Economics, 69(1), 43–55. 10.1016/ j.jue.2010.06.003 Ahn, J., Choi, H., & Oh, D. (2019). Leveraging bridging universities to access international knowl­ edge: Korean universities’ r&d internationalization. Scientometrics, 120(2), 519–537. 10.1007/s111 92-019-03138-4 Ahuja, G. (2000). Collaboration networks, structural holes, and innovation: A longitudinal study. Administrative Science Quarterly, 45(3), 425–455. 10.2307/2667105 Altbach, P. G. (2013). Advancing the national and global knowledge economy: The role of research universities in developing countries. Studies in Higher Education, 38(3), 316–330. 10.1080/03075 079.2013.773222 Appelt, S., Beuzekom, B. van, Galindo-Rueda, F., & Pinho, R. de. (2015). Which factors influence the international mobility of research scientists? In A. Geuna (Ed.), Global mobility of research sci­ entists (pp. 177–213). Elsevier. 10.1016/b978-0-12-801396-0.00007-7 Arrieta, O. A. D., Pammolli, F., & Petersen, A. M. (2017). Quantifying the negative impact of brain drain on the integration of European science. Science Advances, 3(4). 10.1126/sciadv.1602232 Azoulay, P., Ganguli, I., & Zivin, J. G. (2017). The mobility of elite life scientists: Professional and personal determinants. Research Policy, 46(3), 573–590. 10.1016/j.respol.2017.01.002 Balland, P.-A., Boschma, R., & Ravet, J. (2019). Network dynamics in collaborative research in the EU, 20032017. European Planning Studies, 27(9), 1811–1837. 10.1080/09654313.2019.1641187 Barabási, A. L., Jeong, H., Néda, Z., Ravasz, E., Schubert, A., & Vicsek, T. (2002). Evolution of the social network of scientific collaborations. Physica A: Statistical Mechanics and Its Applications, 311(3–4), 590–614. 10.1016/s0378-4371(02)00736-7 Baruffaldi, S. H., & Landoni, P. (2012). Return mobility and scientific productivity of researchers working abroad: The role of home country linkages. Research Policy, 41(9), 1655–1665. 10.1016/ j.respol.2012.04.005 Bathelt, H., & Li, P. (2020). Processes of building cross-border knowledge pipelines. Research Policy, 49(3), 103928. 10.1016/j.respol.2020.103928 Bathelt, H., Malmberg, A., & Maskell, P. (2004). Clusters and knowledge: Local buzz, global pipelines and the process of knowledge creation. Progress in Human Geography, 28(1), 31–56. 10.1191/03 09132504ph469oa Bauder, H. (2015). The international mobility of academics: A labour market perspective. International Migration, 53(1), 83–96. 10.1111/j.1468-2435.2012.00783.x Beaver, D. Deb. (2001). Reflections on scientific collaboration (and its study): Past, present, and future. Scientometrics, 52(3), 365–377. 10.1023/A:1014254214337 Beaver, D. deB, & Rosen, R. (1978). Studies in scientific collaboration. Scientometrics, 1(1), 65–84. 10.1007/bf02016840 Beaver, D. deB, & Rosen, R. (1979). Studies in scientific collaboration part III. Professionalization and the natural history of modern scientific co-authorship. Scientometrics, 1(3), 231–245. 10.1007/ bf02016308 Beine, M., Docquier, F., & Rapoport, H. (2008). Brain drain and human capital formation in developing countries: Winners and losers. The Economic Journal, 118(528), 631–652. 10.1111/j.1468-0297. 2008.02135.x

353

Holger Graf and Martin Kalthaus Beine, M., Docquier, F., & Rapoport, H. (2001). Brain drain and economic growth: Theory and evidence. Journal of Development Economics, 64(1), 275–289. 10.1016/s0304-3878(00)00133-4 Binz, C., & Truffer, B. (2017). Global innovation systems–a conceptual framework for innovation dynamics in transnational contexts. Research Policy, 46(7), 1284–1298. 10.1016/j.respol.2017.05.012 Bozeman, B., & Corley, E. (2004). Scientists’ collaboration strategies: Implications for scientific and technical human capital. Research Policy, 33(4), 599–616. 10.1016/j.respol.2004.01.008 Breschi, S., & Lissoni, F. (2004). Knowledge networks from patent data: Methodological issues and research targets. In H. Moed, W. Glänzel, & U. Schmoch (Eds.), Handbook of quantitative science and technology research: The use of publication and patent statistics in studies on s&t systems (pp. 613–643). Springer. Bush, V. (1945). Science - the endless frontier. A Report to the President by Vannevar Bush, Director of the Office of Scientific Research; Development, July 1945. http://www.nsf.gov/about/history/ vbush1945.htm Cantner, U., & Graf, H. (2011). Innovation networks: Formation, performance and dynamics. In C. Antonelli (Ed.), Handbook on the economic complexity of technological change (pp. 366–394). Edward Elgar. http://www.e-elgar.com/bookentry_main.lasso?id=13391 Cantner, U., & Rake, B. (2014). International research networks in pharmaceuticals: Structure and dynamics. Research Policy, 43(2), 333–348. 10.1016/j.respol.2013.10.016 Carrazza, S., Ferrara, A., & Salini, S. (2016). Research infrastructures in the LHC era: A scientometric approach. Technological Forecasting and Social Change, 112, 121–133. 10.1016/j.techfore.2016.02.005 Catalini, C., Fons-Rosen, C., & Gaulé, P. (2020). How do travel costs shape collaboration? Management Science, 66(8), 3340–3360. 10.1287/mnsc.2019.3381 Cimini, G., Zaccaria, A., & Gabrielli, A. (2016). Investigating the interplay between fundamentals of national research systems: Performance, investments and international collaborations. Journal of Informetrics, 10(1), 200–211. 10.1016/j.joi.2016.01.002 Coccia, M., & Wang, L. (2016). Evolution and convergence of the patterns of international scientific collaboration. Proceedings of the National Academy of Sciences, 113(8), 2057–2061. 10.1073/ pnas.1510820113 Crane, D. (1972). Invisible colleges; diffusion of knowledge in scientific communities. University of Chicago Press. Cronin, B., Shaw, D., & Barre, K. L. (2003). A cast of thousands: Coauthorship and subauthorship collaboration in the 20th century as manifested in the scholarly journal literature of psychology and philosophy. Journal of the American Society for Information Science and Technology, 54(9), 855–871. 10.1002/asi.10278 Cronin, B., Shaw, D., & Barre, K. L. (2004). Visible, less visible, and invisible work: Patterns of collaboration in 20th century chemistry. Journal of the American Society for Information Science and Technology, 55(2), 160–168. 10.1002/asi.10353 Czaika, M., & Orazbayev, S. (2018). The globalisation of scientific mobility, 19702014. Applied Geography, 96, 1–10. 10.1016/j.apgeog.2018.04.017 Defazio, D., Lockett, A., & Wright, M. (2009). Funding incentives, collaborative dynamics and sci­ entific productivity: Evidence from the EU framework program. Research Policy, 38(2), 293–305. 10.1016/j.respol.2008.11.008 Doehne, M., & Rost, K. (2021). Long waves in the geography of innovation: The rise and decline of regional clusters of creativity over time. Research Policy, 50(9), 104298. 10.1016/j.respol.2021.104298 Dopfer, K., Foster, J., & Potts, J. (2004). Micro-meso-macro. Journal of Evolutionary Economics, 14(3), 263–279. 10.1007/s00191-004-0193-0 Dosi, G. (1988). The nature of the innovative process. In G. Dosi, C. Freeman, R. Nelson, G. Silverberg, & L. Soete (Eds.), Technical change and economic theory (pp. 221–238). Pinter. Ebadi, A., & Schiffauerova, A. (2013). Impact of funding on scientific output and collaboration: A survey of literature. Journal of Information & Knowledge Management, 12(04), 1350037. 10.1142/ s0219649213500378 Edler, J., Fier, H., & Grimpe, C. (2011). International scientist mobility and the locus of knowledge and technology transfer. Research Policy, 40(6), 791–805. 10.1016/j.respol.2011.03.003 European Union. (2016). The European Research Area – evolving concept, implementation challenges. European Union. https://www.europarl.europa.eu/thinktank/en/document/EPRS_IDA(2016)579097 Florida, R. (2005). The world is spiky. The Atlantic Monthly, 48–51.

354

Global knowledge embeddedness Fortunato, S., Bergstrom, C. T., Börner, K., Evans, J. A., Helbing, D., Milojević, S., Petersen, A. M., Radicchi, F., Sinatra, R., Uzzi, B., Vespignani, A., Waltman, L., Wang, D., & Barabási, A.-L. (2018). Science of science. Science, 359(6379), eaao0185. 10.1126/science.aao0185 Fox, M. F., Realff, M. L., Rueda, D. R., & Morn, J. (2017). International research collaboration among women engineers: Frequency and perceived barriers, by regions. The Journal of Technology Transfer, 42(6), 1292–1306. 10.1007/s10961-016-9512-5 Frame, J. D., & Carpenter, M. P. (1979). International research collaboration. Social Studies of Science, 9(4), 481–497. 10.1177/030631277900900405 Freeman, L. C. (1979). Centrality in social networks conceptual clarification. Social Networks, 1(3), 215–239. 10.1016/0378-8733(78)90021-7 Freeman, R. B. (2010). Globalization of scientific and engineering talent: International mobility of students, workers, and ideas and the world economy. Economics of Innovation and New Technology, 19(5), 393–406. 10.1080/10438590903432871 Freeman, R. B., Ganguli, I., & Murciano-Goroff, R. (2015). Why and wherefore of increased scientific collaboration. In A. B. Jaffe & B. F. Jones (Eds.), The changing frontier (pp. 17–48). University of Chicago Press. 10.7208/9780226286860-003 Gaillard, J., & Gaillard, A. M. (1997). Introduction: The international mobility of brains: Exodus or circulation? Science, Technology and Society, 2(2), 195–228. 10.1177/097172189700200202 Gazni, A., Sugimoto, C. R., & Didegah, F. (2012). Mapping world scientific collaboration: Authors, institutions, and countries. Journal of the American Society for Information Science and Technology, 63(2), 323–335. 10.1002/asi.21688 Georghiou, L. (1998). Global cooperation in research. Research Policy, 27(6), 611–626. 10.1016/s00487333(98)00054-7 Gibson, J., & McKenzie, D. (2014). Scientific mobility and knowledge networks in high emigration countries: Evidence from the Pacific. Research Policy, 43(9), 1486–1495. 10.1016/j.respol.2014.04.005 Glänzel, W. (2001). National characteristics in international scientific co-authorship relations. Scientometrics, 51(1), 69–115. 10.1023/a:1010512628145 Glänzel, W., & Schubert, A. (2005). Analysing scientific networks through co-authorship. In H. F. Moed, W. Glänzel, & U. Schmoch (Eds.), Handbook of quantitative science and technology research: The use of publication and patent statistics in studies of s&t systems (pp. 257–276). Springer Netherlands. 10.1007/1-4020-2755-9_12 Graf, H., & Kalthaus, M. (2018). International research networks: Determinants of country em­ beddedness. Research Policy, 47(7), 1198–1214. 10.1016/j.respol.2018.04.001 Guan, J., Zhang, J., & Yan, Y. (2015). The impact of multilevel networks on innovation. Research Policy, 44(3), 545–559. 10.1016/j.respol.2014.12.007 Gui, Q., Liu, C., & Du, D. (2019). Globalization of science and international scientific collaboration: A network perspective. Geoforum, 105, 1–12. 10.1016/j.geoforum.2019.06.017 Hidalgo, C. A. (2016). Disconnected, fragmented, or united? A trans-disciplinary review of network science. Applied Network Science, 1(6), 1–19. 10.1007/s41109-016-0010-3 Hoekman, J., Frenken, K., & Tijssen, R. J. W. (2010). Research collaboration at a distance: Changing spatial patterns of scientific collaboration within Europe. Research Policy, 39(5), 662–673. 10.1016/ j.respol.2010.01.012 Hornung, E. (2014). Immigration and the diffusion of technology: The huguenot diaspora in Prussia. American Economic Review, 104(1), 84–122. 10.1257/aer.104.1.84 Hou, L., Pan, Y., & Zhu, J. J. H. (2021). Impact of scientific, economic, geopolitical, and cultural factors on international research collaboration. Journal of Informetrics, 15(3), 101194. 10.1016/ j.joi.2021.101194 Jacob, M., & Meek, V. L. (2013). Scientific mobility and international research networks: Trends and policy tools for promoting research excellence and capacity building. Studies in Higher Education, 38(3), 331–344. 10.1080/03075079.2013.773789 Jeong, S., Choi, J. Y., & Kim, J.-Y. (2014). On the drivers of international collaboration: The impact of informal communication, motivation, and research resources. Science and Public Policy, 41(4), 520–531. 10.1093/scipol/sct079 Jones, B. F. (2009). The burden of knowledge and the “death of the Renaissance man”: Is innovation getting harder? Review of Economic Studies, 76(1), 283–317. 10.1111/j.1467-937x.2008.00531.x

355

Holger Graf and Martin Kalthaus Jones, B. F. (2021). The rise of research teams: Benefits and costs in economics. Journal of Economic Perspectives, 35(2), 191–216. 10.1257/jep.35.2.191 Jones, B. F., Wuchty, S., & Uzzi, B. (2008). Multi-university research teams: Shifting impact, geog­ raphy, and stratification in science. Science, 322(5905), 1259–1262. 10.1126/science.1158357 Jöns, H. (2009). “Brain circulation” and transnational knowledge networks: Studying long-term effects of academic mobility to Germany, 1954–2000. Global Networks, 9(3), 315–338. 10.1111/j.1471-0374. 2009.00256.x Jöns, H. (2007). Transnational mobility and the spaces of knowledge production: A comparison of global patterns, motivations and collaborations in different academic fields. Social Geography, 2(2), 97–114. 10.5194/sg-2-97-2007 Kato, M., & Ando, A. (2017). National ties of international scientific collaboration and researcher mobility found in nature and science. Scientometrics, 110(2), 673–694. 10.1007/s11192-016-2183-z Katz, J. S., & Martin, B. R. (1997). What is research collaboration? Research Policy, 26(1), 1–18. 10. 1016/s0048-7333(96)00917-1 Keller, W. (2004). International technology diffusion. Journal of Economic Literature, 42, 752–782. Kolesnikov, S., Woo, S., Li, Y., Shapira, P., & Youtie, J. (2019). Mapping the emergence of inter­ national university research ventures. The Journal of Technology Transfer, 44(4), 1134–1162. 10. 1007/s10961-017-9640-6 Larivière, V., Gingras, Y., Sugimoto, C. R., & Tsou, A. (2015). Team size matters: Collaboration and scientific impact since 1900. Journal of the Association for Information Science and Technology, 66(7), 1323–1332. 10.1002/asi.23266 Laudel, G. (2002). What do we measure by co-authorships? Research Evaluation, 11(1), 3–15. 10.3152/147154402781776961 Leeuwen, T. N. van. (2009). Strength and weakness of national science systems: A bibliometric analysis through cooperation patterns. Scientometrics, 79(2), 389–408. 10.1007/s11192-009-0426-y Leung, R. C. (2013). Networks as sponges: International collaboration for developing nanomedicine in China. Research Policy, 42(1), 211–219. 10.1016/j.respol.2012.05.001 Luukkonen, T., Persson, O., & Sivertsen, G. (1992). Understanding patterns of international scientific collaboration. Science, Technology, & Human Values, 17(1), 101–126. 10.1177/016224399201700106 Melin, G., & Persson, O. (1996). Studying research collaboration using co-authorships. Scientometrics, 36(3), 363–377. 10.1007/BF02129600 Melkers, J., & Kiopa, A. (2010). The social capital of global ties in science: The added value of international collaboration. Review of Policy Research, 27(4), 389–414. 10.1111/j.1541-1338.2010. 00448.x Moser, P., Voena, A., & Waldinger, F. (2014). German Jewish Émigrés and US invention. American Economic Review, 104(10), 3222–3255. 10.1257/aer.104.10.3222 Narin, F., Stevens, K., & Whitlow, E. S. (1991). Scientific co-operation in Europe and the citation of multinationally authored papers. Scientometrics, 21(3), 313–323. 10.1007/BF02093973 Netz, N., Hampel, S., & Aman, V. (2020). What effects does international mobility have on scientists’ careers? A systematic review. Research Evaluation, 29(3), 327–351. 10.1093/reseval/rvaa007 Newman, M. E. J. (2001). The structure of scientific collaboration networks. Proceedings of the National Academy of Sciences, 98(2), 404–409. 10.1073/pnas.98.2.404 Nguyen, T. V., Ho-Le, T. P., & Le, U. V. (2017). International collaboration in scientific research in Vietnam: An analysis of patterns and impact. Scientometrics, 110(2), 1035–1051. 10.1007/s11192-01 6-2201-1 Niosi, J., & Bellon, B. (1994). The global interdependence of national innovation systems: Evidence, limits, and implications. Technology in Society, 16(2), 173–197. 10.1016/0160-791x(94)90028-0 Ou, A. Y., Varriale, L., & Tsui, A. S. (2012). International collaboration for academic publication: Implications from the resource-based view and transaction cost theory. Group & Organization Management, 37(4), 407–451. 10.1177/1059601112448422 Ozman, M. (2009). Inter-firm networks and innovation: A survey of literature. Economics of Innovation and New Technology, 18(1), 39–67. 10.1080/10438590701660095 Persson, O., Glänzel, W., & Danell, R. (2004). Inflationary bibliometric values: The role of scientific collaboration and the need for relative indicators in evaluative studies. Scientometrics, 60(3), 421–432. 10.1023/B:SCIE.0000034384.35498.7d

356

Global knowledge embeddedness Phelps, C., Heidl, R., & Wadhwa, A. (2012). Knowledge, networks, and knowledge networks: A review and research agenda. Journal of Management, 38(4), 1115–1166. 10.1177/0149206311432640 Picci, L. (2010). The internationalization of inventive activity: A gravity model using patent data. Research Policy, 39(8), 1070–1081. 10.1016/j.respol.2010.05.007 Plotnikova, T., & Rake, B. (2014). Collaboration in pharmaceutical research: Exploration of countrylevel determinants. Scientometrics, 98(2), 1173–1202. 10.1007/s11192-013-1182-6 Pohl, H. (2021). Internationalisation, innovation, and academiccorporate co-publications. Scientometrics, 126(2), 1329–1358. 10.1007/s11192-020-03799-6 Powell, W. W., Koput, K. W., & Smith-Doerr, L. (1996). Interorganizational collaboration and the locus of innovation: Networks of learning in biotechnology. Administrative Science Quarterly, 41(1), 116–145. 10.2307/2393988 Price, D. J. D. S., & Beaver, D. (1966). Collaboration in an invisible college. American Psychologist, 21(11), 1011–1018. 10.1037/h0024051 Ribeiro, L. C., Rapini, M. S., Silva, L. A., & Albuquerque, E. M. (2018). Growth patterns of the network of international collaboration in science. Scientometrics, 114(1), 159–179. 10.1007/s11192017-2573-x Saxenian, A. (2005). From brain drain to brain circulation: Transnational communities and regional upgrading in India and China. Studies in Comparative International Development, 40(2), 35–61. 10. 1007/bf02686293 Scellato, G., Franzoni, C., & Stephan, P. (2015). Migrant scientists and international networks. Research Policy, 44(1), 108–120. 10.1016/j.respol.2014.07.014 Schubert, A., & Braun, T. (1990). International collaboration in the sciences 1981—1985. Scientometrics, 19(1–2), 3–10. 10.1007/bf02130461 Shields, R. (2013). Globalization and international student mobility: A network analysis. Comparative Education Review, 57(4), 609–636. 10.1086/671752 Solla Price, D. J. de. (1963). Little science, big science. Columbia Univ. Press. Teichler, U. (2015). Academic mobility and migration: What we know and what we do not know. European Review, 23(S1), S6–S37. 10.1017/S1062798714000787 Tsouri, M., Hanson, J., & Normann, H. E. (2021). Does participation in knowledge networks facilitate market access in global innovation systems? The case of offshore wind. Research Policy, 50(5), 104227. 10.1016/j.respol.2021.104227 Uzzi, B. (1996). The sources and consequences of embeddedness for the economic performance of organizations: The network effect. American Sociological Review, 61(August), 674–698. 10.2307/ 2096399 Uzzi, B. (1997). Social structure and competition in interfirm networks: The paradox of embedded­ ness. Administrative Science Quarterly, 42(1), 35–67. 10.2307/2393808 Vanni, T., Mesa-Frias, M., Sanchez-Garcia, R., Roesler, R., Schwartsmann, G., Goldani, M. Z., & Foss, A. M. (2014). International scientific collaboration in HIV and HPV: A network analysis. PLoS ONE, 9(3), e93376. 10.1371/journal.pone.0093376 Verginer, L., & Riccaboni, M. (2021). Talent goes to global cities: The world network of scientists’ mobility. Research Policy, 50(1), 104127. 10.1016/j.respol.2020.104127 Verginer, L., & Riccaboni, M. (2020). Cities and countries in the global scientist mobility network. Applied Network Science, 5(1). 10.1007/s41109-020-00276-0 Wagner, C. S., Cai, X., Zhang, Y., & Fry, C. V. (2021). One-year in: COVID-19 research at the international level in CORD-19 data. SSRN Electronic Journal. 10.2139/ssrn.3874974 Wagner, C. S., & Leydesdorff, L. (2005a). Network structure, self-organization, and the growth of international collaboration in science. Research Policy, 34(10), 1608–1618. 10.1016/j.respol.2005.08.002 Wagner, C. S., & Leydesdorff, L. (2005b). Mapping the network of global science: Comparing international co-authorships from 1990 to 2000. International Journal of Technology and Globalisation, 1(2), 185–208. Wagner, C. S., Park, H. W., & Leydesdorff, L. (2015). The continuing growth of global cooperation networks in research: A conundrum for national governments. PLOS ONE, 10(7), e0131816. 10.1371/ journal.pone.0131816 Wagner, C. S., Whetsell, T. A., & Leydesdorff, L. (2017). Growth of international collaboration in science: Revisiting six specialties. Scientometrics, 110(3), 1633–1652. 10.1007/s11192-016-2230-9

357

Holger Graf and Martin Kalthaus Wagner, C. S., Whetsell, T. A., & Mukherjee, S. (2019). International research collaboration: Novelty, conventionality, and atypicality in knowledge recombination. Research Policy, 48(5), 1260–1270. 10.1016/j.respol.2019.01.002 Wagner, C. S., Whetsell, T., Baas, J., & Jonkers, K. (2018). Openness and impact of leading scientific countries. Frontiers in Research Metrics and Analytics, 3(10). 10.3389/frma.2018.00010 Waltman, L., Tijssen, R. J. W., & Eck, N. J. van. (2011). Globalisation of science in kilometres. Journal of Informetrics, 5(4), 574–582. 10.1016/j.joi.2011.05.003 Wang, J., Hooi, R., Li, A. X., & Chou, M. (2019). Collaboration patterns of mobile academics: The impact of international mobility. Science and Public Policy, 46(3), 450–462. 10.1093/scipol/scy073 Wanzenböck, I., Scherngell, T., & Brenner, T. (2014). Embeddedness of regions in European knowledge networks: A comparative analysis of inter-regional R&D collaborations, co-patents and co-publications. The Annals of Regional Science, 53(2), 337–368. 10.1007/s00168-013-0588-7 Wanzenböck, I., Scherngell, T., & Lata, R. (2015). Embeddedness of European regions in European union-funded research and development (R&D) networks: A spatial econometric perspective. Regional Studies, 49(10), 1685–1705. 10.1080/00343404.2013.873119 Wuchty, S., Jones, B. F., & Uzzi, B. (2007). The increasing dominance of teams in production of knowledge. Science, 316(5827), 1036–1039. 10.1126/science.1136099 Youtie, J., Li, Y., Rogers, J., & Shapira, P. (2017). Institutionalization of international university research ventures. Research Policy, 46(9), 1692–1705. 10.1016/j.respol.2017.08.006 Zacchia, P. (2018). Benefiting colleagues but not the city: Localized effects from the relocation of superstar inventors. Research Policy, 47(5), 992–1005. 10.1016/j.respol.2018.03.004 Zitt, M., Bassecoulard, E., & Okubo, Y. (2000). Shadows of the past in international cooperation: Collaboration profiles of the top five producers of science. Scientometrics, 47(3), 627–657. 10.1023/ A:1005632319799

358

27 MACRO-EVOLUTIONARY MODELLING OF CLIMATE POLICIES Karolina Safarzynska

27.1

Introduction

Reducing carbon dioxide emissions has become an urgent necessity. It requires a global transformation of an economic system to limit global warming to 1.5°C (IPCC, 2022). The long-term impacts of climate policies are typically studied with Integrated Assessment Models (IAMs) that capture the co-dynamics of economy and climate (Tol, 1995; Nordhaus, 2017; Weyant, 2017). IAMs use aggregate equations that describe how the accumulation of carbon emissions by production activities affects global temperature, which in turn causes economic damages that reduce economic growth. They rely on the assumption of perfect rationality, representative agents and equilibrium outcomes. There is a large dissatisfaction with the current climate-economy models, coming not only from the heterodox economists, who criticise their core assumptions, but also from prominent mainstream economists, who use such models for climate policy assessment (Pindyck, 2013; Stern, 2013; Weitzman, 2013). The analytical tractability of IAMs comes at the price of unrealistic damage functions. Such models do not account for feedback loops between climate change and agents’ behaviour (Balint et al., 2017). They ignore innovations as inherently uncertain and abstain from financial markets, despite financing investments in renewable energy being recognised as the main obstacle for its wider diffusion (Grubb, 2014; Stern, 2016). Moreover, such models rely on the assumption of full employment, which may blur the distributional impacts of climate policies (Taylor et al., 2016). Amid this criticism, agent-based modelling (ABM) has been suggested as a new wave of economic modelling of climate change impacts (Farmer et al., 2015; Stern, 2016; Arthur, 2021). The method offers a more realistic representation of agents’ heterogeneity, technological innovations, and distributional issues. Such models go beyond a single representative rational agent. Instead, many heterogeneous, boundedly rational agents interact with each other. Agents are described by many behavioural rules that evolve over time (Holland and Miller, 1991). They further can adopt new rules, either by modifying, intentionally or not, existing ones, or by imitating behaviours of other agents who are seen as being successful. Their interactions are characterised by increasing returns, learning, and path dependence.

DOI: 10.4324/9780429398971-30

359

Karolina Safarzynska

ABMs have been successfully used in several disciplines, to study pandemics, bubbles in financial markets, social segregation, or labour markets. Macro-evolutionary models go further by specifying a set of evolutionary mechanisms (diversity, selection and innovation) governing agents’ interactions, and rules regarding R&D activities, income generation, saving-investment decisions, and the evolution of credit connections. Such models are used to study macro phenomena emerging from interactions of boundedly rational agents. They offer a more flexible and realistic characterisation of socio-economic systems than mainstream economics. In this chapter, I discuss contributions of macro-evolutionary modelling to climate policies and formulate suggestions for future research.

27.2

Macro-evolutionary models

Evolutionary theory conceptualises processes of change as the outcome of three processes: variation, selection, and differential replication. Innovation constitutes a mechanism of diversity generation, which can result from series of incremental improvements in already existing technologies or can lead to the emergence of a radically novel solution. Competition, regulation, and institutions act as the main driver of selection reducing diversity in the system. Selection operates primarily by imitation of other actors, thus favouring the diffusion of some options at the cost of others (Nelson and Winter, 1982). Macro-economic models developed in an evolutionary spirit describe the diversity of production techniques at the level of individual firms. In such models, innovation in the driving force behind long-term growth. This relates to the Schumpeter’s idea of creative destruction, where the expected growth rate of the economy depends upon the economywide amount of research. Innovation destroys rents that motivated the previous discovery. Opportunities for innovation can arise at any time, as entities (agents, firms) constantly engage in R&D activities. In the classic evolutionary model of growth by Nelson and Winter (1982, chapter 12), heterogeneous firms produce a homogenous product but with different techniques. Dynamics are driven by investment rules and search processes applied to each individual firm. Nelson and Winter built their evolutionary growth model from the bottomup. They carried out simulations of micro data, which generated patterns consistent with observed macro aggregates. The model initiated a new phase in evolutionary growth theorising. For instance, in the Dosi-type of models (Chiaromonte and Dosi, 1993; Fagiolo and Dosi, 2003; Dosi et al., 2010), the economy is divided into two sectors: the industry fabricating inputs for production and the industry manufacturing final goods. Dynamics at the firm level underlie the growth rate of aggregate output. Instead of costly optimisations, firms develop simple routines, describing investments in R&D activities or in expansion of their production capacity. Other macro-evolutionary studies focus on structural change as a driver of growth. The emergence of new industries, underlying structural change, allows for a continuation of economic development (Saviotti and Pyka, 2004; 2008). New sectors and products emerge due to innovations that build upon previous discoveries, creating path-dependencies in economic development. Because of its explicit micro foundations linked to individual behaviour, evolutionary economics has offered an alternative policy approach to neoclassical economics. This resulted in new contributions to empirical research on patterns of change at the firm, industry, and international levels (Louca, 2020). 360

Macro-evolutionary modelling of climate policies

27.3

Macro-evolutionary modelling of climate change and policies

Recently macro-evolutionary models have integrated energy markets and/or climate cycle to examine interactions between different sub-systems of the economy, necessary to fully understand low-carbon transitions (Gerst et al., 2013; Lamperti et al., 2019; Ponta et al., 2018). Most important contributions have been made in the following areas: (1) to study the role of consumer preferences in diffusion of environmental innovations, (2) in modelling energy transitions, and (3) climate-change finance, which are discussed below (see Balint et al. (2017) and Hafner et al. (2020) offer for more extensive overviews). Early macro-evolutionary models have focused on modelling the substitution of “green” by “dirty” technologies, where the green technology has been defined as having a lower environmental impact than the dirty one. This has been studied with co-evolutionary models of demand and supply dynamics (Windrum and Birchenhall; 1998, 2005; Oltra and Saint-Jean, 2005; Malerba et al., 2008). A basic idea here is that evolving consumers’ preferences affect the direction of firms’ innovative activities. This differs from macroevolutionary growth models, e.g., of the Dosi-type, where demand is presented only at the aggregate level. Instead, in co-evolutionary models, firms engage in product innovations to attract new consumers. For instance, in the model by Windrum and Birchenhall (1998), consumers move between consumer classes depending on the relative attractiveness of products offered by incumbent firms. Evolving preferences of consumers affects the direction of product innovations. Windrum et al. (2009) applied this approach to address the substitution of more by less polluting firms. In their model, firms try to improve characteristics of their products, such as: price, services provided to consumers, and environmental performance. The authors show that successful firms may ignore environmental impacts of their products if this dimension is not perceived by consumers as attractive. As a result, whether a new “clean” technological paradigm emerges depends on the initial distribution of consumers’ preferences. The approach discussed above has been criticised for an unrealistic depiction of environmental innovations (Oltra, 2006). An environmental impact is not a given characteristic but evolves over time, depending on other technologies. For instance, electric cars are often discussed as an example of environmental innovations. However, their environmental impact can be worse than of conventional cars if country’s electricity production relies on coal. Subsequently, macro-evolutionary models have included a more realistic representation of the energy system. For instance, the EURACE-model of the EU economy includes central banks, commercial and investment banks, manufacturing sectors, capital sectors, households in different European countries, which are represented as populations of agents interacting with each other. The model has been extended by the energy sector to study impacts of energy policies (Ponta et al., 2018). In the model, the energy sector includes fossil-fuel and renewable-energy power producers, who make pricing and capacity investment decisions. The results show that the feed-in tariff policy is effective in promoting investments in renewable energy, without undermining government finances. As another example, the ENGAGE model has been proposed to study the impact of international and domestic climate policies (Gerst et al., 2013). The authors extend the model by Dosi et al. (2010) by adding a simplified energy system, where three energy technologies are distinguished: carbon-heavy, carbon-light, and carbon-free. Energy is an input of production in the capital and final goods sectors. The total amount of energy depends on energy intensities

361

Karolina Safarzynska

that are subject to improvements in R&D activities. The authors find that if carbon tax revenues are used to invest in R&D activities of firms, this minimises both the economic and emissions impacts of the tax. There are concerns that improvements in energy efficiency may lead to the rebound effect. The rebound effect describes the phenomenon that energy savings from improvements in energy efficiency are lower than expected due to unintended second-order effects. Safarzynska (2012) proposes a macro-evolutionary model composed of heterogeneous populations of boundedly rational consumers, firms and power producers to study this effect. In the model, the energy sector is composed of heterogonous power plants (nuclear, coal and gas technologies), which technical features have been calibrated on data from the UK electricity system. The type of fuel to be embodied in a new power plant, as well as its size, depends on the energy mix in electricity production at the time of its entry. The author shows that investments in nuclear energy, instead of replacing fossil fuels in electricity generation, may create an additional supply of electricity. In addition, simulations reveal that the rebound effect due to improvements in energy efficiency is greater in industries characterised by the stronger network effect. The latter implies that consumers prefer products that others already have. This may deter, or slow down, the entrance of new firms embodying more energy-efficient technologies, contributing to the rebound effect. Agent-based modelling have proved useful for studying impacts of shocks propagating through the network of interrelated banks and firms (Tedeschi et al., 2012; Thurner and Poledna, 2013). Recently, the approach has been applied increasingly to study financial climate-related risks. Broadly speaking, climate risks can be classified into physical and transition risks. Physical risks capture losses occurring as a result of natural disasters destroying physical capital. Transition risks arise as climate polices induce phasing-out of fossil fuels and scaling up of low-carbon technologies, which can cause revaluation of their assets, affecting balance sheets of financial institutions holding them. As a result, climaterelated shocks can spread through the interbank lending matrix or the network of global financial institutions, compromising financial stability. Battiston et al. (2017) find that over 40% of portfolios of pension funds in the European Union are exposed to devaluation of fossil fuel assets. On contrary to existing climate-economy models, macro-evolutionary models pay attention to financial constraints. Typically, they are stock flow consistent, which implies that a model accounts for all monetary flows in the economy. In such macro-evolutionary models, firms finance their production activities with credit, e.g., capital expansion. If banks have no sufficient liquidity to lend money to a firm, they ask other banks for loans. This way credit connections at the interbank lending market co-evolve with economic activities. This approach turned out useful for studying climate-related risks. In particular, Lamperti et al. (2019) introduce a climate module into the macro-evolutionary model by Dosi et al. (2010). The authors show that climate shocks to labour productivities and capital stocks of individual firms can generate much larger aggregate damages than those obtained by IAMs. Subsequently, the authors estimate total losses due to climate shocks, affecting firms’ profitability and abilities to re-pay loans, on the public costs of bank bailouts. As another example, Safarzynska and van den Bergh (2017a,b) propose an agent-based macroeconomic model to study the impact of energy policies on financial stability. The authors show that in case loans to energy sector concentrate in few banks, the cascade of bankruptcies can spread through the financial network as a result of insolvency of renewable energy producers. 362

Macro-evolutionary modelling of climate policies

27.4

Research gaps and issues for future modelling

Macro-evolutionary economics have made important contributions to improve our understanding of the impacts of climate policies on economic and financial stability. However, there are still important research gaps for future research. This includes: studying how climate policies perform under bounded rationality and social interactions; distributional impacts of climate policies; and the impact of scarcity of non-renewable resources on future growth.

27.5

Climate policy under bounded rationality

As an important area of future applications, macro-evolutionary models can be used to study how energy and climate policies perform under conditions of bounded rationality and social interactions. This is motivated by the fact that climate polices, such as carbon tax, affects all prices in the economy, and thus decisions of boundedly rational consumers, investors and firms. Empirical and experimental research have shown that boundedly rational behaviours such as “satisficing”, habits, imitation, myopia, and heuristics under uncertainty can undermine effectiveness of climate policies. For instance, status consumption have been shown to increase the social cost of carbon (Howarth, 2006), or that ignoring consumers’ myopia may lead to the underestimation of the future carbon dioxide emissions from the transport sector (Safarzynska and van den Bergh, 2018). This relates to fact that “present bias” causes people to undervalue energy costs when they purchase energy-using durables, such as vehicles, refrigerators or air conditioners (Allcott and Wozny, 2014). This make consumers choose energy-inefficient products, with serious repercussions for longterm energy use. Correcting such behaviours could substantially reduce carbon dioxide emissions. For instance, providing households with personalised messages comparing their energy use to other peers can reduce energy consumption by 2% in the United States (Allcott, 2011). Steering behaviours towards sustainability is difficult as they are embedded in a complex network of social institutions and practices that are less accessible for quantitative investigations (Creutzig et al., 2018). Traditional macroeconomic models cannot accommodate heterogeneous agents described by many different heuristics, ignoring diversity of behaviours. This relates to the fact that behavioural utility functions make models analytically intractable and non-linear that an equilibrium may not exist. Macro-evolutionary models are suitable to assess the economy-wide effects of climate policies and how they perform under different forms of bounded rationality and social interactions. Future work in this direction is important, for instance, to study how social comparison and status-seeking may affect what society perceives to be basic necessities, regardless of their environmental impact.

27.6

Distributional issues

The topic of the distributional impacts of climate change and policies have achieved much attention in the literature. In general, carbon taxes are considered to be regressive, meaning that the tax disproportionally burdens low-income households. This raises social concerns and resistance to taxes, which can be illustrated with “yellow vest” protests in France. Integrated Assessment Models are typically used to study the optimal carbon tax at the

363

Karolina Safarzynska

global or regional level (Anthoff et al., 2009; Adler et al. 2017). IAMs are built on a high level of aggregation and thus they do not consider individual consumers. Once the optimal climate policy is determined using such models, its distributional impacts are assessed independently, for instance, using static multi-sector computable general equilibrium (CGE) models or micro-simulation models that make use of microeconomic databases containing data on individual households. However, studying optimal policies together with their distributional consequences is important. The optimal carbon tax depends on economic growth, which in turn is affected by economic inequalities. In addition, studies of distributional impacts of carbon taxes focus on income differences between individuals, while ignoring capital income and unemployment effects of climate change. Yet, differential saving rates from capital and labour income have been argued to be a major driver of inequality (Piketty, 2014). Macro-evolutionary models with a more realistic representation of agents’ heterogeneity have opened up opportunities to explore distributional issues and can be used to address multiple interrelated inequalities. So far, macro-evolutionary theorising and modelling have focused on studying mechanisms through which financialisation of the economy can increase inequality. To illustrate with an example, Russo et al. (2016) propose an agent-based model in which heterogeneous households, firms, and banks interact in the credit market. The authors show that inequality may increase if consumption grows less proportionally than wealth. Consumer credit has two impacts on the economy: it reduces temporarily unemployment while boosting aggregate demand; and it accelerates the system tendency to the crisis. Similarly, in the model by Cardaci and Saraceno (2015), under increasing income inequality, the willingness to lend becomes higher, which may lead to a debt-driven boom and bust cycle. Recently, Botta et al. (2021) show that securitisation and the emergence of complex financial products may lead to a more unequal and unstable economic system. So far, the distributional impacts of climate policies have not achieved much attention in macro-evolutionary studies. As an important exception is Rengs et al. (2020), who propose a macro-evolutionary model that includes different classes of households and two types of innovations aimed at improving either the carbon or labour intensity of production. The results show that a supply-oriented subsidy for green innovation can substantially reduce carbon dioxide emissions without causing negative effects on employment. In future macroevolutionary studies, accounting for different types of inequalities, namely in consumption, wealth and income, and how they affect growth is important to fully capture distributional consequences of climate policies.

27.7

Input-output structure and resource scarcity

During the last century, global materials use increased 8-fold (Krausmann et al., 2009). For the first time in history, the weight of the human-made mass has surpassed the overall living biomass (Elhacham et al., 2020). In many industrial countries, further growth does not improve the quality of life in any significant way (Beddoe et al., 2008), yet it poses an enormous burden on the environment. So far, macro-evolutionary models have ignored resources or material constraints on economic growth. In fact, most macro-evolutionary models typically adopt a linear view on production, where capital constitutes an important input for production of consumer products, but not vice versa. Only recently, there have been attempts to link macroevolutionary models with an input-output structure, where outputs are simultaneously 364

Macro-evolutionary modelling of climate policies

inputs of production in other sectors (Berg et al., 2015; Poledna et al., 2018). In such studies, firms in different manufacturing industries are linked via an input-output matrix. This allows computing physical flows of intermediary inputs in the economy. For instance, in Poledna et al. (2018), the authors model 64 industry sectors composed of heterogeneous firms, who follow simple production heuristics instead of constantly optimising their choices. Each firm in a specific industry employs a fixed coefficient production technology, with the share of inputs in production taken directly from the input-output tables. The model is also linked to national accounts, census data, and business information. It simulates interactions of millions of agents representing each natural person or legal entity in Austria. Such a detailed representation of the national economy allowed the authors to estimate direct and indirect losses due to natural disasters. Next to the input-output structure, accounting for resource scarcity is important for future macro-evolutionary modelling of climate policies. This relates to the fact that production of renewable energy is many times more mineral and metal-intensive compared to fossil fuels. For instance, photovoltaic system uses 11–40 times more copper than fossil fuel generation, while wind power plants use 6–14 times more iron (Hertwich et al., 2014). Today’s mineral supply and investment plans are insufficient for the transformation of the energy sector (IEA, 2022). Many of minerals critical for emerging technologies are in scarcity due to political tensions or shortages (Massari and Ruberti, 2013). According to the World Bank (2020) estimates, global production of critical minerals needed for low-carbon technologies will rise by 200–1000% by 2050. Mineral deposit discovery as well as the grade of processed ore in many metal industries have been declining over time (Michaux, 2021). To fully understand challenges related to scaling up low-carbon technologies, modelling complex interactions between energy, commodity and financial markets is necessary. As an attempt in this direction, Gerdes et al. (2022) propose a macro-evolutionary three-sector model of trade that accounts for mining resources in the Global South. Resources are then used for production of capital goods in the Global North. The authors show that resource extraction and labour conditions in the Global South can affect global emissions from the manufacturing section.

27.8

Conclusions

Macro-evolutionary models relax core assumptions of mainstream climate-economy models. They replace the assumption of market equilibrium and rationality of representative agents with networks of interacting agents, who adapt their behaviour to constantly changing environments. Such models have already provided important insights into climate policies. Future work along these lines is important as the low-carbon transition will take many decades. It requires overcoming bounded rationality of investors, firms and consumers, especially heuristics focused on immediate gratification to support long-term system change. We need more models to help us understand mechanisms underlying such processes. Macro-evolutionary models are especially suitable to study such issues as they pay attention to system connectivity, endogenous technological change and distributional issues.

References Adler, M., Anthoff, D., Bosetti, V., Garner, G., Keller, K., Treich, N., 2017. Priority for the worse-off and the social cost of carbon. Nature Climate Change 7: 443–449.

365

Karolina Safarzynska Allcott, H., 2011. Social norms and energy conservations. Journal of Public Economics 95: 1082–1095. Allcott, H., Wozny, N., 2014. Gasoline prices, fuel economy, and the energy paradox. The Review of Economics and Statistics 96: 779–795. Anthoff, D., Hepburn, C., Tol, R., 2009. Equity weighting and the marginal damage costs of climate change. Ecological Economics 68: 836–849. Arthur, B., 2021. Foundations of complexity economics. Nature Physics 3: 136–145. Balint, T., Lamperti, F., Mandel, A., Napoletano, M., Riventini, A., Sapio, A., 2017. Complexity and the economics of climate change: a survey and a look forward. Ecological Economics 138: 252–256. Battiston, S., Mandel, A., Monasterolo, I., Schütze, F., Visentin, G., 2017. A climate stress-test of the financial system. Nature Climate Change 7(4), 283–288. Beddoe, R., Constanza, R., Farley, J., Garza, E., Kent, J., Kubiszewski, I., Martinez, L., McCowen, T., Murphy, K., Myers, N., Ogden, Z., Stapleton, K., Woodward, J., 2008. Overcoming system roadblocks to sustainability: the evolutionary redesign of worldviews, institutions, and Technologies. PNAS 106, 2483–2489. Berg, M., Hartley, B., Richters, O., 2015. A stock-flow consistent input-output model with applications to energy price shocks, interest rates and heat emissions. New Journal of Physics. 015011. Botta, A., Caverzasi, E., Russo, A., Gallegati, M., Stiglizt, J., 2021. Inequality and finance in a rent economy. Journal of Economic Behavior and Organization 183: 998–1024. Cardaci, A. Saraceno, F., 2015. Inequality, financialisation and economic crises: an agent-based macro model. Working paper 2015-27, OFCE. Chiaromonte, F., Dosi, G., 1993. Heterogeneity, competition and macroeconomic dynamics. Structural Change and Economic Dynamics 4: 36–63. Creutzig, F., Roy, J., Lamb, W.F. et al., 2018. Towards demand-side solutions for mitigating climate change. Nature Climate Change, 8: 260–263. Dosi, G., Fagiolo, G., Roventini, A., 2010. Schumpeter meeting Keynes: a policy-friendly model of endogenous growth and business cycles. Journal of Economic Dynamics and Control 34: 1748–1767 Elhacham, E., Ben-Uri, L., Grozovski, J., Bar-On, Y.M., Milo, R., 2020. Global human-made mass exceeds all living biomass. Nature 588: 442–444. Fagiolo, G., Dosi, G., 2003. Exploitation, exploration and innovation in a model of endogenous growth with locally interacting agents. Structural Change and Economic Dynamics, 14, 237–273. Farmer, J.D., Hepburn, C., Mealy, P., Teytelboym, A., 2015. A third wave in the economics of climate change. Environmental Resource Economics 62: 329–357. Gerdes, L., Rengs, B., Scholz-Wackerle, M., 2022. Labour and environment in global value chains: an evolutionary policy study with a three-sector and two-region agent-based macroeconomic model. Journal of Evolutionary Economics 32: 132–173. Gerst, M.D., Wang, P., Roventini, A., Fagiolo, G., Dosi, G., Howarth, R.B., Borsuk, M.E., 2013. Agent-based modeling of climate policy: An introduction to the ENGAGE multi-level model framework. Environmental Modelling & Software 44: 62–75. Grubb, M., 2014. Planetary Economics. Routledge, Oxford and New York. Hafner, S., Anger-Kraavi, A., Monasterolo, I., Jones, A., 2020. Emergence of new economics energy transition models: A review. Ecological Economics 177: 106779. Hertwich, E.G., Gibon, T., Bouman, E.A., 2014. Integrated life-cycle assessment of electricity-supply scenarios confirms global environmental benefit of low carbon technologies. PNAS 112: 6277–6282. Holland, J.H., Miller, J.H. 1991. Artificial adaptive agents in economic theory. American Economic Review 81: 365–370. Howarth, R.B., 2006. Optimal taxes under relative consumption effects. Ecological Economics 58: 209–219. IEA, 2022.The role of Critical Minerals in Clean Energy Transitions. IPCC, 2022. Impacts, Adaptation and Vulnerability. The Six Assessment Report. Krausmann, F. Gingrich, S., Eisenmenger, N., Erb, K-H., 2009. Growth in global materials use, GDP and population during the 20th century. Ecological Economics. Lamperti, F., Bosetti, V., Roventini, A., Tavoni, M., 2019. The public costs of climate-induced financial instability. Nature Climate Change 9(11): 829–833. Louca, F., 2020. Chris Freeman forging the evolution of evolutionary economics. Industrial Corporate Change 29: 1037–1046.

366

Macro-evolutionary modelling of climate policies Malerba, F., Nelson, R., Orsenigo, L., Winter, S., 2008. Public policies and changing boundaries of firms in a “history friendly model” of the coevolution of the computer and semiconductor industries. Journal of Economic Behaviour and Organisation 67: 355–380. Massari, S., Ruberti, M., 2013. Rare earth elements as critical raw materials: Focus on international markets and future strategies. Resources Policy 38(1), 36–43. Michaux, S., 2021. The Mining of Minerals and the Limits to Growth. Geological Survey of Finland Report. Nelson, R.R., Winter, S.G., 1982. An Evolutionary Theory of Economic Change. Harvard University Press, Cambridge, MA. Nordhaus, W.D., 2017. Revisiting the social cost of carbon. PNAS 114: 1518–1523. Oltra, O., 2006. Environmental innovation and industrial dynamics: the contributions of evolutionary economics. Mimeo. Oltra, O., Saint-Jean, M., 2005. The dynamics of environmental innovations: three stylised trajectories of clean technology. Economics of Innovation and New Technology 14: 189–212. Piketty, 2014. Capital in the Twenty-First Century. Cambridge, MA., Belknap Press. Pindyck, R.S., 2013. Climate change policy: what do the models tell us? Journal of Economic Literature 51: 860–872. Poledna et al. 2018 (many authors). When does a disaster become a systemic event? Estimating indirect economic losses from natural disasters. Mimeo. Ponta, L., Raberto, M., Teglio, A., Cincotti, S., 2018. An agent-based stock-flow consistent model of the sustainable transition in the energy sector. Ecological Economics 145: 274–300. Rengs, B., Scholz-Wäckerle, M., van den Bergh, J., 2020. Evolutionary macroeconomic assessment of employment and innovation impacts of climate policy packages. Journal of Economic Behavior and Organization 169: 332–368. Russo, A., Riccetti, L., Gallegati, M., 2016. Increasing inequality, consumer credit and financial fragility in an agent based macroeconomic model. Journal of Evolutionary Economics 25: 25–47. Safarzynska, K., 2012. Modeling the rebound effect in two manufacturing industries. Technological Forecasting and Social Change 79: 1135–1154. Safarzynska, K., 2017a. The implications of industrial development for fuel diversification. Ecological Economics 137: 37–46. Safarzynska, K., van den Bergh, J., 2017b. Integrated crisis-energy policy: macro-evolutionary modeling of interactions between technology, finance and energy systems. Technological Foresting and Societal Change 114: 119–137. Safarzynska, K., van den Bergh, J., 2018. A higher rebound effect under bounded rationality: Interactions between car mobility and electricity generation. Energy Economics. 74: 179–196. Saviotti, P.P., Pyka, A., 2004. Economic development by the creation of new sectors. Journal of Evolutionary Economics 14: 1–35. Saviotti, P.P., Pyka, A., 2008. Micro and macro dynamics: industry life cycles inter-sector coordination and aggregate growth. J Evolutionary Economics 18: 167–182 Stern, N., 2013. The structure of economic modeling of the potential impacts of climate change: grafting gross underestimation of risk onto already narrow science models. Journal of Economic Literature 51: 838–859. Stern, N., 2016. Current climate models are grossly misleading. Nature 530: 407–409. Taylor, L., Rezai, A., Foley, D.K., 2016. An integrated approach to climate change, income distribution, employment, and economic growth. Ecological Economics 121: 196–220. Tedeschi, G., Mazloumian, A., Gallegati, M., Helbing, D. 2012. Bankruptcy Cascades in Interbank Markets. PLOS ONE, 7(12): e52749. Thurner, S., Poledna, S. 2013. DebtRank-transparency: Controlling systemic risk in financial networks. Scientific Reports 3: 1888. https://doi.org/10.1038/srep01888 Tol, R.S.J. 1995. The damage costs of climate change toward more comprehensive calculations. Environmental and Resource Economics 5: 353–374. Weitzman, M.L., 2013. Tail-hedge discounting and the social cost of carbon. Journal of Economic Literature 51: 873–882. Weyant, J., 2017. Some contributions of integrated assessment models of global climate change. Review of Environmental Economics and Policy 11: 115–137.

367

Karolina Safarzynska Windrum, P., Birchenhall, C.T., 1998. Is life cycle theory a special case?: dominant designs and emergence of market niches through co-evolutionary learning. Structural Change and Economic Dynamics 9: 109–134. Windrum, P., Birchenhall, C.T., 2005. Structural change in the presence of network externalities: a coevolutionary model of technological successions. Journal of Evolutionary Economics 15: 123–148. Windrum, P., Carli, T., Birchenhall, C., 2009. Environmental impact, quality, and price: consumer tradeoffs and the development of environmentally friendly technologies. Technological Forecasting and Societal Change 76, 533–551. World Bank 2020. Minerals for climate action: The mineral intensity of the clean energy transition. World Bank.

368

28 THE VISIBLE HAND OF INNOVATION POLICY Uwe Cantner and Claudia Werker

28.1

Introduction

While artificial intelligence (AI) has been a major game changer in technological, economic, and societal development (OECD, 2019), questions on how agency and power are dis­ tributed between human and artificial intelligence have not been addressed conclusively so far (Willson, 2016). A case in point is innovation policy, because – when it comes to AI – innovation policy focuses on implementing the usual measures leaving any ethical questions to expert councils (OECD, 2019). The problem with this line of action is that ethical questions involve decisions on values, i.e. “… things worth striving for” (Taebi, Correljé, Cuppen, Dignum, & Pesch, 2014, p. 119) and even more so on shared values which require to integrate the values of all relevant stakeholders (Werker, 2021). And, it is completely unclear how experts can be sufficiently legitimized to decide on shared values. In contrast, policy makers do have the legitimacy to define values and shared values, because they represent the elected government and because they bring together various stakeholders ei­ ther in formal or informal settings. Innovation policy measures targeting AI have been widely adopted in the OECD countries (see for this and the following OECD, 2019). Yet they have been mostly motivated by the systems approach. Innovation policy from a systemic perspective deals with system failures such as missing innovative agents, relationships, or institutions (e.g. Edquist, 2011; Klein Woolthuis, Lankhuizen, & Gilsing, 2005). Using the system failures approach for AI means that measures have included the stimulation of AI research, the fostering of AI talent, the support of the development and adoption of AI solutions as well as of AI-driven businesses (OECD, 2019). While the systems failure approach and related policy broadly acknowledge and address problems stemming from the emergence and use of AI, we suggest that it insufficiently captures three issues constituting AI as a game changer for innovation policy. Each of these issues leads to a research question: The first issue is that innovation policy makers have been outsourcing ethical issues to expert councils (see first paragraph of introduction). Accordingly, we pursue the following research question:

DOI: 10.4324/9780429398971-31

369

Uwe Cantner and Claudia Werker

RQ1How can policy makers ensure that all stakeholders involved in and affected by AI, i.e. being legitimized to have a say about innovation processes and outcomes, contribute to developing shared values that guide technological progress of AI? The second issue is that AI is not only a technology with one trajectory to exploit but emerges as a method of invention (Cockburn, Henderson, & Stern, 2018). Therefore, there is not only the question of how much to exploit a specific technological trajectory of AI but also the question of how to change trajectories to explore more promising routes – and in fact how to define what promising routes are. Accordingly, we pursue the following research question: RQ2How can policy makers know how to influence the intensity and the direction of AI, i.e. how to implement the shared values? The third issue in need of attention is deep learning, i.e. combining different elements of machine learning where the understanding or prediction of the world takes place without any further human intervention (Taddy, 2018). This lack of human involvement in deep learning means that not only innovation policy is many steps away from having any say in its development but so are other stakeholders in the innovation system. Accordingly, we pursue the following research question: RQ3How can policy makers ensure that the human factor is sufficiently embedded in deep learning of AI? To answer these questions we create a concept of innovation policy that uses three elements: (1) We start from the stakeholders who are legitimized in the process and result of innovation (de Saille, 2015). (2) We integrate this with the approach of Schumpeterian catalytic innovation policy (Cantner & Vannuccini, 2018) which is “… capable to maneuver the parallel necessities to influence both the intensity and the direction of innovative activities” (Cantner & Vannuccini, 2018, p. 834). (3) Moreover, we use the Responsible Research and Innovation (RRI) systems approach which is able to systematically involve all stakeholders in AI processes (Werker, 2021). The chapter is organized as follows: We start by introducing three essential elements of AI. Then, we discuss the technology AI as a game changer for innovation policy. After that, we introduce the visible hand of innovation policy at the interface of artificial and human intelligence. We conclude with our main insights and draw up a couple of major open research questions emerging from them.

28.2

Three essential elements of AI

Intelligence displayed by humans is the starting point when it comes to defining AI. Human intelligence emerges from learning and problem solving (Gardner, 1999). The capabilities of human intelligence serve as benchmark to assess the abilities of artificial intelligence. The weakest version of AI has only limited application areas, stronger versions might match human abilities and the strongest ones would even go beyond them (Kaplan & Haenlein, 2019). Generally speaking, AI can be defined “… as a system’s ability to interpret external data correctly, to learn from such data, and to use those learnings to achieve specific goals and tasks through flexible adaptation” (Kaplan & Haenlein, 2019, p. 17). Full end-to-end AI 370

The visible hand of innovation policy

solutions are able to absorb human-level knowledge such as machine reading, thereby carrying out tasks previously done by human beings (cf. this and the following Taddy, 2018). They require three essential elements. 1 Machine learning (ML): ML routines are able to detect patterns and make predictions. Simple ML algorithms analyze historical data and are therefore “… basically limited to predicting a future that looks mostly like the past” (Taddy, 2018, p. 2). 2 A domain structure which mirrors the context of your AI by breaking complex problems into composite tasks. This domain structure relies heavily on domain expertise, e.g. in a business setting on business and economic expertise. This expertise provides the “rules of the game”. As long as they are clear the tasks can be solved with ML. With the help of a domain structure combinations of ML algorithms can be used, i.e. “… dynamic processes designed and implemented by humans in conjunction with technical affordances and within broader political, social and cultural environments that are shaped by the con­ tinual interactions of strategies, structures and tactics” (Willson, 2016, p. 148). 3 Data generation, i.e. “… steady stream of new and useful information flowing into the composite learning algorithms” (Taddy, 2018, p. 4). This is not simply data collection but a strategy to develop the massive bank of data required to get the system up and running and to keep producing data so that the system can learn. Data generation in this sense requires the use of big data and internet of solutions (Kaplan & Haenlein, 2019). One of AI’s elements, i.e. ML, shows the typical features of a general purpose technology (GPT). (cf. this and the following Taddy, 2018). ML “… in its current form has become a general purpose technology. These tools are going to get cheaper and faster over time, due to innovations in the ML itself and above and below in the AI technology stack” (Taddy, 2018, p. 3). So, ML shows all three defining features of general purpose technologies, i.e. pervasiveness, innovation spawning, and scope for improvement (Helpman & Trajtenberg, 1994). These three character­ istics of a GPT are not given by nature but the result of an ongoing mechanism of interaction and further innovation, the so-called dual inducement mechanism (Bresnahan & Trajtenberg, 1995). Given a sector supplying a GPT and a number of application sectors the dual inducement mechanism runs between the two types of sectors as follows: An increase in the quality of the GPT (the so-called “technological dynamism”) incentivizes the actors in the application sectors to increase their technological level (the “innovation complementarities” property of GPTs), and this, in turn, induces the GPT sector to advance its technology, and so forth. The dual inducement mechanism entails two features (Bresnahan & Trajtenberg, 1995). First, it connects various actors and therefore provides for breadth in the application of the GPTs’ technological core; this, in a technological sense, contributes to an alignment of respectively widespread innovation activities. Secondly, the mechanism induces a direction of innovation activities and offers orientation by precluding alternative ways of further develop­ ment; this, in a behavioral sense, aligns innovation activities and reduces some of the uncer­ tainty inherent to innovation activities. Both features on the one hand provide for efficiency in innovation activities and contribute to their intensity – with all benefits to income and welfare. On the other hand, however, over time both features tend to build up dependency and to precluding alternative options and paths of development. A technological dynamism that turns out to be efficiency enhancing in the beginning over time may be causal for inflexibility in choice because of presumably high costs of switching to an alternative – the typical feature of so-called lock-in situations (Arthur, 1989; Cantner & Vannuccini, 2017). 371

Uwe Cantner and Claudia Werker

28.3

AI as a game changer for innovation policy

AI is inherently different from other technologies because it comes with three major is­ sues: (1) Innovation policy systematically outsources the core questions regarding developing shared values for AI to expert councils. (2) AI goes way beyond a general purpose technology (GPT). (3) AI involves deep learning which endangers human involvement in the exploitation and exploration of its potential. We consider these issues in the following.

28.4 28.4.1

Legitimacy of stakeholders and ethical guidelines for AI

Legitimacy of stakeholders and outsourcing of ethical issues of AI to expert councils

The legitimacy of stakeholders – often represented by the political actors – to intervene in innovation is nicely summarized by market and system failures. These failures have to do with the very process and context under which innovations are generated and introduced. This perspective, however, leaves out the very characteristics of the innovations generated and the very impact those innovations have on social and environmental structures and settings (e.g. are they environmentally friendly or not, are they health friendly or not, do they have socially positive or negative consequences?). Hence, when we talk about the legitimacy of stakeholders/policy makers to intervene, we consider this not only – quite traditionally – related to processes but also to outcomes of activities in general and of innovation activities in particular. On this basis, three arguments can be put forward that justify the legitimacy to intervene when innovation outcomes and their relation to values are concerned. This legitimacy can be attributed, first, to a shift in perspective of the relation between the sphere of science and the sphere of society in the following sense (de Saille, 2015). In a Polanyi world, science (as a major source of innovation) takes place in a neutral space (republic of science) where political, moral, and social questions and hence values do not play a role. This neutralization and separation meanwhile is not considered proper any­ more. Science is rather seen as embedded in the political, social, and economic world with causations running both ways and where values are relevant. Secondly, this legitimacy arises from a prominent concept, Schumpeter (1942/1975) “creative destruction” due to which any new idea and innovation has also destructive effects which affect values – and we observe this today in the social as well as the natural/environmental domains. Third, legitimacy can be justified also in an intergenerational context (Cantner & Vannuccini, 2018). Were future generations able to negotiate with the concurrent generations about which new ideas and innovations to implement quite some innovation directions would not have taken the way which in the end has been pursued – the Friday-for-Future initiative suggests such a con­ nection. Politicians today could serve as the attorney of these future generations and in doing so need to have on board proper values – the German “Klimakabinett” could be interpreted in this way. Along with these three arguments, proper policy approaches move away from a topdown government to a more reciprocal structure of governance (cf. this and the following de Saille, 2015). Herein policy has been moving away from considering other stakeholders as being ignorant towards respecting their questions as legitimized value-based questions about technological development. 372

The visible hand of innovation policy

Policy makers seem to be aware of the ethical issues coming with AI (cf. this and the following OECD, 2019). Yet, they ignore the question of legitimacy of stakeholders, because in the OECD countries, they outsource ethical questions of values to expert councils. We suggest that this is not the right way to deal with ethical AI issues, because those legitimized to take action to deal with AI issues do not take responsibility.

28.4.2

Ethical guidelines of expert councils have numerous problems

AI ethics initiatives in the form of expert councils come with numerous problems. First, they are very broad and simplistic while not giving much practical guidance on how to deal with real-world issues emerging around AI (cf. this and the following Copeland, 2019). They only help for rather simple human expert hand-craft machine learning models but do not cover any advanced AI solutions where AI algorithms decide what factors are relevant or where ML models iterate rapidly based on constantly incoming data streams. Moreover, AI ethics initiatives “… ignore fundamental normative questions about what kind of society we want” (D’Ignazio & Klein, 2019) and instead focus “… on more procedural technical and legal concerns …” (Kitchen, 2019). The concerns about outsourcing ethical AI issues to expert councils become particularly clear when following the reasoning of Mittelstadt (2019) who compares AI with the medical community. He suggests that in contrast to the medical developments AI development lacks (1) common aims and fiduciary duties, (2) professional history and values, (3) proven methods to translate principles into practice, and (4) robust legal and professional accountability mechanisms. … We must therefore hesitate to celebrate consensus around high-level principles that hide deep political and normative dis­ agreement. Shared principles are not enough to guarantee ‘Trustworthy AI’ or ‘Ethical AI’ in the future. Without a fundamental shift in regulation, translating principles into practice will remain a competitive, not cooperative, process. (Mittelstadt, 2019) We therefore suggest that while AI is an extremely difficult technology to be accompa­ nied by policy measures, it lies without any doubts within the core responsibility of inno­ vation policy. The oversimplified ethical rules of expert councils suggest that AI solutions are designed carefully in advance giving ample room to detect, discuss and overcome possible negative effects. However, as the Collingridge control dilemma for rapidly evolving technologies, such as AI, has shown, negative effects are often only known after the technologies have been in full use. The dilemma runs thus: ‘attempting to control a technology is difficult … because during its early stages, when it can be controlled, not enough can be known about its harmful social consequences to warrant controlling its development; but by the time these consequences are apparent, control has become costly and slow’ (Collingridge, 1980: 19). … alongside these explicit references to his work, Collingridge’s thinking also has implicit influence on RRI – for instance in the recognition of the significance of corrigibility in the form of ‘responsiveness’. Stilgoe et al. (2013: 1572) describe this as the ‘capacity to change shape or direction in response to stakeholder and public values and changing circumstances’. (Genus & Stirling 2018, p. 63) 373

Uwe Cantner and Claudia Werker

28.5

AI is more than a general purpose technology

AI comprises not only ML, i.e. the first element mentioned in the previous section, but relies heavily on the two other elements that turn it into something “intelligent”, i.e. domain expertise to build a domain structure and constantly instreaming high-quality data (cf. this and the following Taddy, 2018). Particularly, full end-to-end AI solutions require experts who can break complex human problems into composite tasks which ML can solve. And this goes way beyond computer games in which ML are often applied to test their power. Currently, many big data applications still use human expert hand-craft machine learning models, i.e. they still rely heavily on frequent human input (Copeland, 2019). AI as a whole is often addressed as GPT (Cockburn et al., 2018). The fact that AI inherently changes the way innovation processes are carried out, i.e. in its exploitation AI emerges as a method of invention (Cockburn et al., 2018), shows that it is a GPT. Consequently, it reduces uncertainty and increases efficiency. However, this comes at a cost, namely in terms of decreasing flexibility in the direction of innovation activities and their increasing dependency on the GPT. Lock-in situations can arise and then become severe problems when a need to change direction in order to exploit new innovation opportunities by giving up well-known terrain becomes increasingly costly. We suggest that AI goes well beyond a GPT with broad applicability, because it is agenda setting for ongoing as well as future innovation activities. Under these cir­ cumstances, the need to change directions because the current ones are exploited is a lesser problem compared to the following one: The need to change may relate to the fact that outcomes of innovation activities do not meet socially acceptable criteria anymore. Overcoming such situations is a problem of collective action depending on the distribution of preferences and attitudes and hence the heterogeneity of actors in being affected by stalemate situations that are socially unacceptable. For overcoming that and for inducing a self-organizing process of leaving a lock-in, political inter­ vention may be a proper solution (Cantner & Vannuccini, 2017). In this sense, inno­ vation policy has not only to deal with the intensity of innovative change and common lock-in situations resulting from the fact that current directions are technologically exploited. It also has to deal with lock-in situations that mirror socially non-acceptable situations.

28.6 Deep learning driving AI endangers human involvement in decision processes As AI is agenda setting but might not mirror socially acceptable solutions the challenge is to figure out whether and how the direction AI is taking and developing is sufficiently transparent and manageable by stakeholders including policy makers based on shared values. AI has been driven by machine learning. Recently, it has been shifting from “only” machine learning solutions, i.e. understanding or predicting the world based on historical experience, towards actual deep learning, i.e. combining different elements of machine learning where the understanding or prediction of the world takes place without any further human intervention (Taddy, 2018). This lack of human involvement in deep learning means that not only innovation policy is many steps away from having any say in its development but so are all stakeholders in the innovation system.

374

The visible hand of innovation policy

28.7

Using the visible hand of innovation policy at the interface of human and artificial intelligence 28.7.1

The visible hand ensuring the legitimacy of stakeholders

In order to use the visible hand of innovation policy to ensure the legitimacy of stake­ holders, we suggest to follow a number of the suggestions made by (Mittelstadt, 2019), in particular: 1 to create accountability, implementation, and review structures at the organizational level and the sectoral level 2 to include stakeholders case-by-case, i.e. not to develop overarching principles but start bottom-up in developing ethical rules for specific cases 3 to consider establishing AI development as a profession such as medical doctors or lawyers 4 to install ethical principles on the organizational rather than on the individual level, i.e. ethics of business practices and 5 to see ethical guidelines as a process guiding technology development as it evolves.

28.7.2 The visible hand changing direction in AI development To use the visible hand of innovation policy to change directions in AI development we suggest addressing the lock-in situations emerging from AI mirroring socially nonacceptable situations. Addressing AI as a GPT – even when using a system approach as in the OECD countries (OECD, 2019) – would lead to subsidizing activities emerging from it in rather unspecific ways. Yet the crucial question here is what kind of measures policy makers can in principle implement to change the directions within a technology develop­ ment. The related question of timing and of how to determine why and when such a change might be necessary will be discussed below in the context of how to involve human intel­ ligence in the process. In the context of a “usual” GPT the question of how to implement redirection measures has been already answered by the Schumpeterian catalytic R&I policy approach. This approach is … capable to manoeuvre the parallel necessities to influence both the intensity and the direction of innovative activities. In other words, to know when to intervene on the incentive to exploitation of given technological trajectories, and when to intervene easing the transition from an exploited technological trajectory to others, richer in opportunities (hence, on the incentive to exploration). (Cantner & Vannuccini, 2018, p. 834)! The Schumpeterian catalytic R&I policy can be summarized by providing a broad defi­ nition. First, it is called Schumpeterian in order to emphasize that this type of policy is not to be considered a repair shop restoring the incentives of private actors to innovate; it is rather considered a means to push forward specific innovative solutions for (pressing) problems by creating new markets and thereby redirecting and activating private entrepreneurs. These new markets are needed to redirect the innovation activities from a known and established innovation trajectory that is not desirable anymore – as in the case of exploited 375

Uwe Cantner and Claudia Werker

technological potentials or – more relevant in this chapter – as with AI innovation outcomes that do not meet the criterion of social consensus. The term catalytic is used to label this type of policy style since policy-making should intervene in the domain of innovative activities as a catalyst intervenes in a chemical reaction. In contrast with the “market creation” approach, a catalytic public intervention is less persistent; it intervenes directly with its “visible hand”, but it is smart enough to retreat its hand when the “reaction” leading to the enhanced innovative activities in new directions reaches the self-sustaining threshold. In a sense, a catalytic R&I policy is a form of “balancing” intervention: Policy-making should focus on the framework conditions that can favor the establishment of “critical masses” (Witt, 1997) of choice in one or the other direction. New directions offer economic agents to autonomously engage in the exploitation of new or other opportunities that have the potential to meet social consensus. These critical masses can be reached through a temporary direct intervention (e.g. through innovative public procurement), or by helping to define the blurring boundaries of competition between alternative directions. By this, intervention implies that public policy has to adopt sophisticated criteria to discriminate between alternatives in the context of uncertainty and potential overall social consensus. The Schumpeterian catalytic R&I policy shows features that go beyond being Schumpeterian and being catalytic. It furthermore is situation-sensitive, as it combines a “continuity” rationale – justified by the presence of challenges to policy interventions into the innovation realm that remains stable and persistent over time – with a “discontinuity” rationale – motivated by the specific trends of innovative activities in a given historical period (e.g. societal needs, grand challenges, lack of social consensus as in case of certain AI developments). The policy further is experimental, as – within its rationale and given a rather broad “mission” – it should create alternative competing arenas and platforms with new potential for innovation that may achieve a social consensus. The experimental nature of catalytic R&I policy is particularly important, as it subsumes one important dimension related to innovative activities: The incentive for a (guided) bottom-up self-discovery (Foray, 2013; Hausmann & Rodrik, 2006). Self-discovery can be conceived as a criterion for action that has the potential to compensate and mild the risk of governmental failures, as the role of the public is to design the mechanism easing the directional exploration of new trajectories. The design of the experimental arena itself is key to the success of R&I policy. Last but not least, the policy approach is wary, as it has to be built on the awareness that even a limited intervention may lock-in the system into inferior technologies, standards, and social values. Lock-ins are rarely irreversible in the real world (Cantner and Vannuccini, 2017), but the costs deriving from the inflexibility they generate have to be kept lower than the benefits of directional exploration. The case of a lock-in when lacking social consensus – as potentially in the AI case – is concerned is not straight forward. Whenever the costs to switch to another technological trajectory or opportunity are higher than the costs related to staying in one currently addressed which includes the costs accruing from not having a social consensus such a lock-in situation can come up.

28.7.3 The visible hand as guardian of human involvement in the era of deep learning To use the visible hand as a guardian of human involvement in the era of deep learning we use an approach focussing on jointly developing values about the process and outcomes of 376

The visible hand of innovation policy

I. identifying structural components: including all stakeholders and their values as well as ensuring a basic digital literacy of all parties involved

II. finding crucial processes including those to identify shared values

III. assessing components and processes based on shared values with the help of big data analysis controlling for privacy and security issues, fair welfare distribution, strategic behaviour and biases

V. feeding back big data and loT based solutions for problems based on shared values into 1. u. 2.

IV. deriving (valuerelated) drivers and bottlenecks of desirable processes with the help of big data and loT solutions

Figure 28.1

A scheme for assessing RRI systems (see Werker, 2021, p. 278).

research and innovation by including all stakeholders, also those not directly involved in innovation decisions (Werker, 2021). In fact, this approach follows the call for integrating the values of all relevant stakeholders. (European_Commission, 2013; Taebi et al., 2014). A systematic involvement of all stakeholders as suggested by the RRI systems approach ap­ pears to be giving a practical answer to how to come to shared values (Werker, 2021). Citizen science, hence the involvement of the society in invention and innovation process, could be interpreted in this sense. As such the RRI systems approach is closely related to the mission-oriented policy which stresses the importance of the “development of social capabilities, coordinate initiatives and public-private partnerships, foster synergies, and promote the introduction of new combinations that create Schumpeterian rents” (Mazzucato & Penna 2016, 316). An RRI system contains “all relevant stakeholders of RRI and the way their values affect their activities, relationships and supporting institutions” (Werker, 2021, p. 304). In Figure 28.1, you find a “… scheme capturing the structural elements and the processes of innovation systems comprising the following five steps”: i identifying the structural components, i.e. innovative agents, relationships, and institu­ tions (Klein Woolthuis et al., 2005) 377

Uwe Cantner and Claudia Werker

ii finding crucial activities, such as knowledge advance by and diffusion amongst stake­ holders, entrepreneurial experimentation, legitimation, market formation, development of institutions, and influence on the direction of search by different selection mecha­ nisms, such as business models, technology development, market and institutional forces (Bergek et al., 2008; Edquist, 2011) iii assessing components and processes by uncovering desirable ones (Bergek et al., 2008; Edquist, 2011; Klein Woolthuis et al., 2005) iv deriving drivers and bottlenecks of desirable components and processes (Bergek et al., 2008; Edquist, 2011; Klein Woolthuis et al., 2005) v feeding back solutions for problems into the structural components (I.) and processes (II.) including their functioning and co-evolution (Werker, 2021, p. 305f). In the course of AI development, all stakeholders have to get some understanding of how the collection of large amounts of data, often life data, can take place and how this data can be used in AI solutions (Werker, 2021). Those stakeholders less digitally educated than other stakeholders in an RRI system will most likely fall behind (Sogeti, 2013). This opens ample opportunity for governmental, academic and civic agents to step up by educating and involving these disadvantaged stakeholders (see step III in Figure 28.1). As big data has been driving big science, i.e. data-driven solutions in research, e.g. at CERN (Sogeti, 2013), we might expect that the values emerging from, in this case, the academic sector, might already include goals of inclusiveness and enabling people by educating them. As long as an RRI system is not dominated by profit-oriented organizations only there is a good chance that the RRI process will lead to shared values providing a level playing field in the RRI system. (Werker, 2021, p. 314) Using AI applications all stakeholders have to be aware of privacy and security issues as well as concerns regarding welfare, discrimination, and strategic behavior. After having determined the structural components and processes of the RRI system it is crucial to address this potential issue in the assessment of them in steps III to V of Figure 28.1. Dealing with the lack of human involvement in AI requires an approach that goes beyond traditional approaches. AI’s potential can only unfold by providing structure and rules around messy business scenarios (Taddy, 2018) and societal processes at large, not only including agents from the industrial but also from the governmental and scientific sectors. As such, AI might not only form the problem but also (parts) of the solutions to human involvement in decision processes, particularly regarding monitoring processes. AI might help in collecting and potentially including the values of all stakeholders affected by AI solutions. On this background, we suggest to use the responsible research and innovation (RRI) systems approach, because involving all parties in an RRI process is at its very heart (cf. this and the following Werker, 2021). Particularly, innovation policy in RRI systems aims at developing shared values of innovative agents actively carrying out RRI as well as of sta­ keholders who are only subject to their effects (Taebi et al., 2014). Generally speaking, this approach is well suited to deal with the additional complications and opportunities of digital transformation. It helps to point at and to deal with privacy issues as well as the risk of discrimination and manipulation severely increasing in the digital age. Moreover, it 378

The visible hand of innovation policy

shows how big data analytics and Internet of Things solutions offer multiple opportunities of following RRI processes more closely, thereby offering chances to sufficiently integrate shared values in the RRI.

28.8

Conclusions

AI and the algorithms involved in its use … invoke questions about how to conceptualise issues such as agency and power within a technologised everyday. … Science and Technology Studies, software studies and actor network theory all provide some fruitful insights and methods particularly in relation to the specificity of particular algorithms, yet largely fail to address many of the broader issues and questions of the everyday that are raised. (Willson, 2016, p. 148) In this chapter, we addressed these broader issues and questions related to AI, partic­ ularly questions on how agency and power are distributed between human and artificial intelligence. We suggest using the visible hand of innovation policy in three ways when dealing with AI: (1) by involving clearly legitimized stakeholders in the design of ethical guidelines – and avoiding outsourcing this important task to expert councils; (2) by using policy measures that can distinguish between exploration and exploitation of AI; and (3) by a coordinated approach of involving stakeholders in several steps ensuring the implemen­ tation of their shared values in AI-driven decision processes. Such an approach can neither rely on policy actions nor market relationships alone but has to acknowledge their joint use (Etzkowitz, 2006). While our integrated approach highlights three major issues and principle ways of dealing with them, so far it is still not clear how innovation policy as suggested can exactly take place. Guidelines as subsumed when discussing the involvement of legitimized stake­ holders might help implement a bottom-up case-by-case approach when it comes to designing ethical guidelines for AI. Following a catalytic Schumpeterian approach, as suggested, will help change trajectories within AI, thus changing gear from exploration to exploitation. And the scheme provided in Figure 28.1 is a first step towards guarding human involvement in the era of deep learning. Combined with the ethical guidelines it might lead to a way of using AI while at the same time implementing shared values in all RRI systems of a society.

References Arthur, B. W. (1989). Competing technologies, increasing returns, and lock-in by historical events. The Economic Journal, 99(394), 116–131. Bergek, A., Jacobsson, S., Carlsson, B., Lindmark, S., & Rickne, A. (2008). Analyzing the functional dynamics of technological innovation systems: A scheme of analysis. Research Policy, 37(3), 407–429. doi: 10.1016/j.respol.2007.12.003 Bresnahan, T. F., & Trajtenberg, M. (1995). General purpose technologies ‘Engines of growth’? Journal of Econometrics, 65(1), 83–108. doi: 10.1016/0304-4076(94)01598-T Cantner, U., & Vannuccini, S. (2017). Innovation and lock-in. In H. Barthelt, P. Cohendet, P. L. Simon, & S. Henn (Eds.), The Elgar Companion to Innovation and Knowledge Creation (pp. 165–181). Cheltenham: Edward Elgar.

379

Uwe Cantner and Claudia Werker Cantner, U., & Vannuccini, S. (2018). Elements of a Schumpeterian catalytic research and innovation policy. Industrial and Corporate Change, 27(5), 833–850. doi: 10.1093/icc/dty028 Cockburn, I. M., Henderson, R., & Stern, S. (2018). The Impact of Artifical Intelligence on Innovation. Cambridge: NBER. Collingridge, D. (1980). The Social Control of Technology. London: Pinter. Copeland, E. (2019). Does the public sector really need a code of AI ethics. Retrieved from https:// www.nesta.org.uk/blog/does-public-sector-really-need-code-ai-ethics/ D’Ignazio, C., & Klein, L. (2019). Introduction. Data Feminism. de Saille, S. (2015). Innovating innovation policy: The emergence of ‘Responsible Research and Innovation’. Journal of Responsible Innovation, 2(2), 152–168. doi: 10.1080/23299460.2015.1045280 Edquist, C. (2011). Design of innovation policy through diagnostic analysis: Identification of systemic problems (or failures). Industrial and Corporate Change, 20(6), 1725–1753. doi: 10.1093/icc/dtr060 Etzkowitz, H. (2006). The new visible hand: An assisted linear model of science and innovation policy. Science and Public Policy, 33(5), 310–320. European_Commission. (2013). Options for Strengthening Responsible Research and Innovation. Luxembourg: European Commission. Retrieved from ISBN 978-92-79-28233-1. Foray, D. (2013). The economic fundamentals of smart specialisation. Ekonomiaz, 83(2), 83–102. Gardner, H. (1999). Intelligence Reframed: Multiple Intelligences for the 21st Century. New York: Basic Books. Genus, A., & Stirling, A. (2018). Collingridge and the dilemma of control: Towards responsible and accountable innovation. Research Policy, 47(1), 61–69. doi: 10.1016/j.respol.2017.09.012 Hausmann, R., & Rodrik, D. (2006). Doomed to choose: Industrial policy as predicament. Retrieved from Helpman, E., & Trajtenberg, M. (1994) A Time to sow and a time reap: Growth based on general purpose technologies. In: Vol. 4854. NBER Working Paper Series (pp. 45). Cambridge: NBER. Kaplan, A., & Haenlein, M. (2019). Siri, Siri, in my hand: Who’s the fairest in the land? On the inter­ pretations, illustrations, and implications of artificial intelligence. Business Horizons, 62(1), 15–25. doi: 10.1016/j.bushor.2018.08.004 Kitchen, R. (2019). The ethics of smart cities. Retrieved from https://www.rte.ie/brainstorm/2019/ 0425/1045602-the-ethics-of-smart-cities/ Klein Woolthuis, R., Lankhuizen, M., & Gilsing, V. (2005). A system failure framework for inno­ vation policy design. Technovation, 25(6), 609–619. doi: 10.1016/j.technovation.2003.11.002 Mazzucato, M., & Penna, C. C. R. (2016). Beyond market failures: The market creating and shaping roles of state investment banks. Journal of Economic Policy Reform, 19(4): 305–326. doi: 10.1080/ 17487870.2016.1216416 Mittelstadt, B. (2019). AI Ethics – Too Principled to Fail? SSRN, 15. Retrieved from https://papers. ssrn.com/sol3/papers.cfm?abstract_id=3391293 OECD. (2019). The Digital Innovation Policy Landscape in 2019. Schumpeter, J. A. (1942/1975). Capitalism, Socialism and Democracy. New York: Harper & Row. Sogeti. (2013). No More Secrets with Big Data Analytics. Retrieved from Groningen: http://www.google. de/url?sa=t&rct=j&q=&esrc=s&frm=1&source=web&cd=1&ved=0CD0QFjAA&url=http%3A%2F %2Fvint.sogeti.com%2Fwp-content%2Fuploads%2F2013%2F11%2FSogeti_NoMoreSecrets.pdf& ei=6zqwU4WvJImw7AbysoHQCg&usg=AFQjCNFMAmRVE6v4loN3Kjsg7Vp_sBguKg&bvm= bv.69837884,d.bGE Stilgoe, J., Owen, R., & Macnaghten, P. (2013). Developing a framework for responsible innovation. Research Policy, 42(9), 1568–1580. Taddy, M. (2018). The Technological Elements of Artificial Intelligence. Cambridge: NBER Taebi, B., Correljé, A., Cuppen, E., Dignum, M., & Pesch, U. (2014). Responsible innovation as an endorsement of public values: The need for interdisciplinary research. Journal of Responsible Innovation, 1(1), 118–124. doi: 10.1080/23299460.2014.882072 Werker, C. (2021). Assessing responsible research and innovation (RRI) systems in the digital age. In E. Yaghmaei & I. Van de Poel (Eds.), Assessment of Responsible Innovation: Methods and Practices. Abingdon: Taylor & Francis. Willson, M. (2016). Algorithms (and the) everyday. Information, Communication & Society, 20(1), 137–150. doi: 10.1080/1369118x.2016.1200645 Witt, U. (1997). “Lock-in” vs. “critical masses” — Industrial change under network externalities. International Journal of Industrial Organization, 15(6), 753–773. doi: 10.1016/S0167-7187(97)00010-6

380

29 GENERALIZED RULES, NELSON–WINTER ROUTINES, AND OSTROM RULES Georg D. Blind

29.1

Introduction

Rules are the devices we use to build and transform societies and their economic systems. Explicit or implicit, general or special, shared by many or few, rules inform the cognition and the behavior of agents. Empirically, the processing of rules causes a great many of the phenomena that scholars of human behavior are interested in. Wherever sociologists identify patterns of conduct or social practices, psychologists inquire into habits, and economists observe preferences being revealed, rules of various kinds are at play. We posit that two of the most prominent strands of research in evolutionary economics, namely the works of Nelson–Winter and Elinor Ostrom, may analytically be conceived in terms of a generalized rule approach proposed by Dopfer (2001, 2004, 2005) and summarized in Dopfer and Potts (2008). At first sight, Nelson–Winter’s discussion of routines and Elinor Ostrom’s frequent references to strategies and internalized norms seem to suggest that their analytical units cannot possibly be conceived of as rules. Yet, we hold that terminological choices are history-bound: meaning originates from a constantly evolving context. In essence, this implies that terminology may be considered much less stable than analytical substance. As we shall reason later when delving into their analytical substance, such “terminological fluctuation” may explain why Nelson–Winter argued “routines” in their conceptual works (see also Blind 2016), and why Elinor Ostrom chose to label “rules” only some of the analytical units in her studies of resource governance systems. In spite of their being contemporaries, cross-references between Nelson–Winter and Elinor Ostrom are scarce and of a rather general nature (Blind 2016). Arguably a consequence of their strong focus on very different research objects and their steadily growing followership, such as looking beyond their respective fields, was not an urgently felt need. Yet, in one of her last manuscripts, Elinor Ostrom eventually addressed the need for a generalization of analytical approaches (Ostrom and Basurto 2011): “If we are to […] develop a general theory of institutional change, we must widen our view and study a much more diverse set of rule systems”. To enable advances of magnitudes similar to those by Nelson–Winter and Ostrom for further research areas in economics, we trust that generalizing what we may refer to as “deductive format” in a first step will prove pertinent.

DOI: 10.4324/9780429398971-32

381

Georg D. Blind

Among the attempts at developing a theoretical corpus for evolutionary economics, the rule-centered works of Dopfer (2001, 2004, 2005) and Dopfer and Potts (2008) stand out for their rigorous level of abstraction. Having since earned substantial acclaim (e.g., as “a very interesting approach” in Ostrom and Basurto 2011: 333), empirical application has been sketched (Blind and Pyka 2014), tested (Blind 2012; Grebel 2013; Wäckerle 2013; Scheltjens 2015; Blind 2017), and routes to formalization have been made available (Dosi and Nelson 2018: 81–83). Section 29.2 briefly sketches the Generalized Rule Approach of Dopfer and Potts. We then assess congruity with the seminal approaches of Nelson–Winter (Section 29.3) and E. Ostrom (Section 29.4). Section 29.5 discusses the findings, and Section 29.6 concludes.

29.2 The generalized rule approach and rules as deductive formats The Generalized Rule Approach builds on the rule-centered works of Dopfer (2001, 2004, 2005) and has first been fully presented by Dopfer and Potts (2008). An overview by the same authors (2009) summarizes its essentials. Key to understanding the analytical worldview of the Generalized Rule Approach is its distinction between economic growth and economic evolution. Distinguishing operations from the rules that they are informed by, helps to draw a critical analytical line between the “commodity space” and the “knowledge space”. Economic evolution, thus, is tantamount to the evolution of knowledge understood as a change in the variety and distribution of rules as deductive formats. The unit of change is a trajectory uniting the mechanisms of origination, selective adoption, and retention of a novel rule. Here, “mechanisms” are to be understood as concepts describing the processes by which a novel rule comes about, and eventually gets selected by and retained in a “rule carrier”. The trajectory captures evolution as a diffusion process during which a growing population of agents adopts a novel rule. Dopfer and Potts choose to universally denote as “rules” all deductive formats in their theory. As the meaning of that term is arguably history-bound and context-dependent, let us briefly look at some of the meanings that it may encompass. Originally indicating a physical device for taking measures (e.g., a “folding rule”), it acquired an abstract reading as “authority”, where a “ruler” and later “the rule of law(s)” and other systems are understood to measure (assess) conditions and prescribe corresponding action. Three characteristics may be drawn from this: (1a) a rule requires a measurable condition (1b) the prescription of corresponding operations (2) governing requires a multitude of rules. From (1) we further understand how rules serve as deductive formats, or “condition-action statements” (Dosi and Nelson 2018: 81). Adopting this generalized definition, it becomes obvious that regulation or law (as in Brennan and Buchanan 1985) is but one type of rule. As building block for economic analysis, we argue that rules equally need to embrace routines (Nelson and Winter 1982; see Section 29.3), strategies and norms (Ostrom and Basurto 2011: 321), habits and institutions (Veblen 1909: 628; also see Blind 2016), custom, norms and conventions (Weber 1980[1922]), and even religious belief or myth (Blind and Steineck 2020). To adequately accommodate this universe of deductive formats, Dopfer and Potts propose a taxonomy of four classes and three orders of rules (2008: 6–10). Importantly, the 382

Generalized rules, Nelson–Winter routines, and Ostrom rules

original purpose of Dopfer and Potts is not to generalize deductive formats as such, but plainly to enable a consistent analytical handling of economics subject-matters. Thus, their taxonomy of classes (“types”) and orders (“hierarchy”) should be understood as an invitation to consider the many commonalities of deductive formats used in various areas of economics research. Dopfer and Potts’ four classes refer to common properties of rules that allow a consistent analytical “handling” across research areas. In Dopfer and Potts, behavioral and cognitive rules refer to deductive formats in individual subjects. Referred to as “subject rules” they inform the behavior and understanding of agents, informing their “internal organization”. In turn, social and technical rules govern the arrangement of humans into organizations and of physical objects into artifacts. Referred to as “object rules”, they are understood to guide human and the design of technology. The orders of rules in Dopfer and Potts’ Generalized Rule approach are meant to conceive hierarchy as a structural property of socio-economic systems. Center-stage are their 1st order rules that can be understood as directly informing operations in an economic system. Their 2nd order rules pertain to change in individual rules or ensembles thereof, such as a high “tolerance of failure” moderates the prevalence of innovation based on trial-and-error. And ultimately, 0th order rules are “the set of legal, political, social and cultural rules, or ‘constitutive rules’, that define what is possible and permissible in the economic system” and “define the ‘opportunity space’ of permissible 1st order operations” (Dopfer and Potts 2008: 14–15). For organizing the analytical treatment of rules and operations in various dimensions, Dopfer, Foster and Potts (2004) have introduced a three-layer architecture of micro-mesomacro. For the analysis of rules, this helps to consistently conceive of rules, individual agents, and populations of agents, as well as the entire rule systems. In concrete terms, a rule originates in one agent in the micro domain, diffuses into a population of agents in the meso domain, and eventually causes a change in the rule structure of the macro domain. In contradistinction, at the level of operational analysis, extant rules are used for operations and the dynamic is not a dynamic of rules but rather one of operations based on given rules. We have an initial operation (or rule use) in the micro domain, the frequency of operations in the meso domain (e.g., production volumes), and structural change through changing patterns of operations based on a given rule structure in the macro domain (Blind and Pyka 2014: 1087). In their generalized rules approach, Dopfer and Potts’ predominant concern is a consistent analytical “handling” of the knowledge foundation of the economy. Considered by many as the true missing piece of economic theory, the remainder of this manuscript commits to verifying whether the Generalized Rule Approach succeeds in providing a consistent analytical reading of the milestone approaches of Nelson–Winter and Elinor Ostrom. If it eventually does, we hold that it bears tremendous relevance for the future of economic science.

29.3 Nelson–Winter routines as outward expression of social rules Can we find congruity between the classes and orders in the Generalized Rule Approach and Nelson–Winters analytical architecture? – We start by briefly reviewing major analytical terms in Nelson–Winter’s works to test their congruity with the rule classes of the Generalized Rule Approach: social and technical rules as “object rules”, and behavioral and cognitive rules as “subject rules”. 383

Georg D. Blind

Borrowing their own words, Nelson–Winter’s “real concern is with organizations” (1982: 72). To start with, their routines qualify as social rules (defined as rules guiding the arrangement of individuals into organizations), and more specifically, as rules that have reached the third diffusional stage of retention. As we have presented in the preceding section, Dopfer and Potts define social rules as one of the two classes of rules in the category of object rules. The second class in this category is technological rules, which can be understood as blueprints for arranging physical objects into technologies. Thus, congruity with Nelson–Winter’s reasoning can hardly be disputed for either of these two classes of rules. But can congruity equally be found for the pair of behavioral and cognitive classes of “subject rules”? – Nelson–Winter actually provide an important cue to this question where they note: “Individual skills are the analogue of organizational routines” (1982: 73; and similar in 2002: 30). Transferring their analogy to Dopfer and Pott’s architecture of rules, we find that a “skill” is the outward “expression” of a behavioral rule in its third diffusional stage of retention. And indeed, in Nelson–Winter’s definition of skills as “units of purposive behavior” and as “programmatic” we find further proof that their “skills” qualify as “behavioral rules”. Finally, “cognitive rules” as the fourth class of rules in the Generalized Rule Approach is a term not directly used in Nelson–Winter. But where they refer to “deliberation” (1982: 69), the Generalized Rule Approach identifies “cognitive rules” as guiding that deliberation. Such guidance is necessary, because deliberation is required in cases where agents are confronted with incomplete information, contingency, and fragmented decisions (1982: ibid). In the Generalized Rule Approach, cognitive rules are part of the category of “subject rules”. This is obviously so because cognition is exclusive to individual subjects, not objects, including organizations and “representative agents”. Nelson–Winter only rarely refer to “rules”. As a matter of fact, the term “rule” did not even figure on a list of alternatives: “habit”, “plan”, “script”, and “program” (1982: 74). We may only muse about their motivations, but one possible reason may be their focus on the “outward expression” of rules, namely on skills and routines. In Dopfer and Potts’s reading, such outward expressions are labeled “actualizations” of rules retained by carriers: skills in individuals, and routines in organizations. Looking at the terms separately, we find that Nelson–Winter are using “skill” for labeling programmatic behavior. We may thus suggest to more precisely label the program a behavioral rule, and relegate “skill” to the (routinized) ability to conduct operations by applying a rule. In a similar vein, it may enhance analytical clarity to understand Nelson–Winter’s organizational capabilities as the outward manifestation of social rules (rules for arranging individuals into organizations). Table 29.1 in Table 29.1 Rule Classes in Nelson–Winter, Ostrom, and the Generalized Rule Approach Classes of Rules Subject Rules Nelson–Winter Ostrom

Decision rule (e.g., investment rule) Strategies

Dopfer–Potts

Cognitive rule

Object Rules

Skill Internalized norms Behavioral rule

384

Organizational routine; control mechanisms Seven types of rules

Technology

Social rule

Technical rule

-

Generalized rules, Nelson–Winter routines, and Ostrom rules Table 29.2 Orders of Rules in Nelson–Winter, Ostrom, and the Generalized Rule Approach Orders of Rules Nelson–Winter Ostrom (1990 terminology) Dopfer–Potts

“Institutional matters” Boundary rules (constitutional choice rules) 0th order constitutional rules

Procedure rules Operational rules; position rules; payoff rules 1st order operational rules

Search rules Aggregation rules (collective choice rules); information rules 2nd order mechanism rules

Section 29.5 summarizes terminological congruency for the classes of rules in Nelson–Winter and in the Generalized Rule Approach. Building on Cyert and March (1963) Nelson–Winter also discuss a hierarchy of rules (1982: 17) similar to the three orders of rules in the Generalized Rule Approach. Starting from Dopfer and Potts’ “0th order constitutional rules”, we find these simply referred to as “institutional matters” in Nelson–Winter’s most prominent writings (1982: 363), receiving particular attention only the studies on “national innovation systems” (e.g., in Nelson 1993). The emphasis of their work is clearly on the outward manifestations of 1st and 2nd order rules. For the sake of simplicity, we only referred to instances of 1st order rules in our above discussion of congruity between the classes of rules in the Generalized Rule Approach and Nelson–Winter’s works. However, 2nd order rules are equally present in Nelson–Winter’s writings where they refer to “search rules” (1982: 19), that they link to change and innovation in social systems. In more concrete terms, Nelson–Winter note: “people within the firm may engage in scrutiny of what the firm is doing and why it is doing it, with the thought of revision or even radical change”. Most notably, Nelson–Winter designate these processes as “rule-guided” and “assume a hierarchy of decision rules with higher-order procedures” (1982: 17). Table 29.2 in Section 29.5 summarizes terminological congruency for orders of rules in Nelson–Winter and the Generalized Rule Approach.

29.4

Deductive formats in E. Ostrom’s analytical architecture

As is commonly known, Elinor Ostrom’s “real concern” is about communities. In her view, all rules ultimately exist for a social purpose: “All rules are the result of […] efforts to achieve order and predictability among humans” (Ostrom et al. 1994: 38). In her early writings Ostrom distinguished seven “classes” of rules (1994: chapter 2), but later decided to refer to “types” rather than “classes” (2011: 323). As we shall see, this seemingly minor change is instrumental to evidencing congruency with the rule architecture of the Generalized Rule Approach. Ostrom’s seven types of rules are all aimed at the coordination of individuals in organizations. In the generalized rule approach, this characteristic implies to understand them as “social rules”. In concrete terms, Ostrom (2011: 323) distinguishes: • • • • •

position rules describing conditions and rights for a position in a social system, boundary rules regulating entry to and exit from the system, choice rules prescribing choice conditions for specific positions, aggregation rules specifying voting processes, information rules indicating transparency levels for specific positions, 385

Georg D. Blind

• pay-off rules controlling the distribution of rents, and • scope rules specifying quantitative limitations of operations where monitoring of actions is difficult. In addition to what we thus may label Ostrom’s seven types of social rules, she also discusses two other of the four rules classes of the Generalized Rule Approach, albeit without referring to them as “rules”. Instead, Ostrom applies the terms “strategy” and “norm” (2011: 321). If we mirror her definition of “strategies” as “plans made by individuals in a situation as to what actions they plan to undertake so as to achieve outcomes given their information about the basic structure of the situation” (ibid), we understand that this equals the basic definition of rules as condition-action-statements in the Generalized Rule Approach. And as “plans” her “strategies” are internal to the minds of agents. Hence, they qualify as “cognitive rules”. “Norms”, on the other hand, are seen by Ostrom as “prescriptions about actions or outcomes”, and mostly “are acquired in the context of a community in which the individual frequently interacts” (2011: 322). As she defines her “norms” as “prescriptions”, we understand that they equally qualify as rules. The likely origin within a community that she cites, further indicates that her “norms” are congruent to the class of behavioral rules in the Generalized Rule Approach. Arguably owing to Ostrom’s empirical interests, we cannot identify any discussion of technology in terms of “technical rules” in her writings. Table 29.1 in Section 29.5 summarizes terminological congruency for the classes of rules in Ostrom’s works and in the Generalized Rule Approach. While Ostrom does not formally distinguish “orders” of rules, she implicitly recognizes a hierarchy where she refers to a “constitutional choice-level” (2011: 326) and “operationallevel systems” (ibid. 329). As seen from the Generalized Rule Approach, Ostrom’s abovenoted “boundary rules” (or constitutional choice rules in her 1990 terminology) and “information rules” directly correspond to 0th order rules. Another four of her “types of social rules” can be considered 1st order operational rules; as Ostrom alludes to herself by specifying her rule types for an “operational-level” system (2011: 329). And indeed, much of her work was focusing on how various resourcemanagement systems are structured and governed. We may thus argue that her focus was on the functioning, and less on the evolution of these systems. Still, Ostrom has minutely studied a great number of change processes in resourcemanagement systems. Through an empirical lens, every case is very specific, particularly for dynamics involved in processes of change. But her works eventually do discuss “rules for changing rules” (1990: 202), which clearly corresponds to the 2nd order (change) mechanism rules of the Generalized Rule Approach. In the above list of seven rule types, it is “aggregation rules” that clearly qualify as 2nd order rules through their regulating of voting processes for bringing about change in a social system.1 Table 29.2 in Section 29.5 summarizes terminological congruency for orders of rules in Ostrom’s works, and in the Dopfer–Potts Generalized Rule Approach.

29.5

Discussion

In the Generalized Rule Approach, Nelson–Winter’s “decision rules” are adequately conceptualized as cognitive rules, and their “organizational routines” and “control mechanisms” can be fully embraced in terms of social rules. While congruency seems obvious in these cases, 386

Generalized rules, Nelson–Winter routines, and Ostrom rules

our arguing that Nelson–Winter’s “skills” qualify as behavioral rules might require closer inspection. We believe that congruency is given for three specific reasons. Firstly, behavioral rules are equally understood as units of programmatic behavior that consciously and subconsciously guide economic operations (1982: 74 et seq.). Secondly, Nelson–Winter acknowledge the multi-dimensional heterogeneity of the generalized rule approach: the repertoire of skills (rules) retained differs between agents, and the choice of rules for operations by an individual agent depends on the social context. And thirdly, the repertoire in Nelson–Winter agents is subject to change, that is, it evolves. Elinor Ostrom’s analytical architecture inductively originated from her famous empirical works. Late in her career, though, she raised the need for a more general view on “rule systems”: “we must widen our view and study a much more diverse set of rule systems” (2011: 335). And she went on designating Dopfer and Potts’ Generalized Rules “a very interesting approach” for that purpose (2011: 333). When assessing Ostrom’s analytical devices through the lens of that “interesting approach”, all her “seven types of rules” correspond to social rules, her “strategies” pertain to the class of cognitive rules, and her “internalized norms” to the class of behavioral rules. Reflected in her admitting to the need for a “more diverse set of rule systems”, her works do not include technical rules. Both Nelson–Winter’s and Ostrom’s approaches to modeling the dynamics of social systems acknowledge a hierarchy of rules, but only the Generalized Rule Approach ventures into referring to such “levels” more formally as “orders”. As our assessment has shown, Nelson–Winter’s and Ostrom’s works do reflect on the three orders of rules suggested by the Generalized Rule Approach, albeit with a differing focus. Ostrom’s works generally center on the functioning of specific communities in a system of 1st order operational rules with the institutional frame of 0th order rules and 2nd order rules for formalized change largely given. In contrast, Nelson–Winter’s particular interest in innovation has their focus lean toward the various workings of 2nd order search rules in an operational 1st order system.

29.6

Conclusion

Can the Generalized Rule Approach with its architecture of four classes and three orders of rules embrace the full analytical depth of Nelson–Winter’s and Ostrom’s works? – As presumptuous as it may sound, the assessment presented here implies that this question may be answered in the positive. At the same time, the Generalized Rule Approach can by no means be thought of as a potential substitute. This is simply because generalized rules themselves do not contribute theoretical specifications for firms or resource-management systems. This being so, by design, implies that the Generalized Rule Approach needs to be specified for their respective areas of interest. But if bolstering milestone works like Ostrom’s and Nelson–Winter’s magna opera by reformulating even only parts thereof in terms of the formally more rigorous Generalized Rule Approach, advances in analytical reach can be added to their legacy. In concrete terms, Nelson–Winter’s study of innovation can be brought to the study of institutional selftransformation without any loss of analytical rigor. For instance, the Generalized Rule Approach makes explicit the distinction between rules and corresponding operations that is still partly implicit in both Nelson–Winter’s and Ostrom’s writings. It also fully generalizes processes of change and employs heterogeneous agents open to learning. It thus represents a fully general framework for “the study of the evolution of human societies” as envisioned by 387

Georg D. Blind

Ostrom (2011: 333). In this sense, the high level of generalization in the Dopfer–Potts approach will enable to reap true “epistemological value added”. Moreover, our hopes are that the Generalized Rule Approach will help to alleviate the fragmentation observed in heterogeneous economics. This is because it enables research on subjects beyond innovating firms and resource-management systems with a consistent and formally more rigorous analytical approach. Embracing heterogeneity between and within agents and economic systems and their evolution will likely never come with the “convenience” many economists continue to depend on. But drawing on Generalized Rules as a formally consistent analytical approach, it will at least become feasible to inquire about the many subjects that have been neglected so far. Elinor Ostrom herself once argued that “no one can legislate a language for a scientific community” (1986: 5). And this was arguably never the purpose behind the development of the Generalized Rules Approach. In essence, what Dopfer and Potts have offered, is an invitation to use their theoretical framework with its common terminology for accommodating and formally pushing further the many insights brought about by the community of economists not following the mainline course.

Note 1 In Ostrom’s earlier terminology two rule types were prone to create confusion: “constitutional choice rules” and “collective choice rules” (1990). Her later re-labeling of the former into “boundary rules” (0th order constitutional rules) and the latter into “aggregation rules” (2 order (change) mechanism rule) has eliminated the risk of such confusion.

References Blind, G. 2012. “Entrepreneurial spirit in Japan: An investigation using the rule-based approach”. Evolutionary and Institutional Economic Review 9(2): 183–198. Blind, G. 2016. “Behavioral rules: Veblen, Nelson-Winter, Oström and beyond”. In Frantz, Chen, et al. (Eds.), Handbook of Behavioral Economics. Milton Park: Routledge, pp. 139–151. Blind, G. 2017. “The Entrepreneur in Rule-based Economics: Theory, empirical practice, and policy design”. Economic Complexity and Evolution (Series). Cham: Springer International Publishing. Blind, G. and A. Pyka. 2014. “The rule approach in evolutionary economics: A methodological template for empirical research”. Journal of Evolutionary Economics 24(5): 1085–1105. DOI: 10.1007/s00191014-0382-4. Blind, G. and R. Steineck. 2020. “The missing piece in E. Cassirer’s theory of symbolic forms: The economy”. Evolutionary and Institutional Economic Review (18): 291–315. DOI: 10.1007/s40844-02 0-00191-0. Brennan, G. and J. M. Buchanan. 1985. The Reason of Rules. Cambridge: Cambridge University Press. Cyert, R. M., and J. G. March. 1963. A Behavioral Theory of the Firm. Englewood Cliffs: Prentice-Hall. Dopfer, K. 2001. “Evolutionary economics – Framework for analysis”. In: K. Dopfer (Ed.), Evolutionary Economics: Program and Scope. Boston, Dordrecht, London: Kluwer Academic Publishers, 1–44. Dopfer, K. 2004. “The economic agent as rule maker and rule user: Homo Sapiens Oeconomicus”. Journal of Evolutionary Economics 14: 177–195. Dopfer, K. 2005. “Evolutionary economics: A theoretical framework”. In: K. Dopfer (Ed.), The Evolutionary Foundations of Economics. Cambridge: Cambridge University Press, pp. 3–55. Dopfer, K. and J. Potts. 2008. The General Theory of Economic Evolution. London: Routledge. Dopfer, K., J. Foster, and J. Potts 2004. “Micro-meso-macro”. Journal of Evolutionary Economics 14(3): 263–279.

388

Generalized rules, Nelson–Winter routines, and Ostrom rules Dosi, G. and R. Nelson. 2018. “Technological advance as an evolutionary process”. In: R. Nelson et al. (Ed.), Modern Evolutionary Economics. Cambridge University Press, pp. 35–84. Grebel, T. 2013. “On the tradeoff between similarity and diversity in the creation of novelty in basic science”. Structural Change and Economic Dynamics 27(0): 66–78. Nelson, R. R. (ed.) 1993. National Innovation Systems: A Comparative Analysis. Oxford: Oxford University Press. Nelson, R. R. and S. G. Winter. 2002. “Evolutionary theorizing in economics”. Journal of Economic Perspectives 16(2): 23–46. Nelson, R. R. and S. Winter. 1982. An Evolutionary Theory of Economic Change. Cambridge: Harvard University Press. Ostrom, E. 1990. Governing the Commons. Cambridge: Cambridge University Press. Ostrom, E., R. Gardner and J. Walker 1994. Rules, Games, and Common-Pool Resources. Ann Arbor: University of Michigan Press. Ostrom, E. and X. Basurto. 2011. “Crafting analytical tools to study institutional change”. Journal of Institutional Economics 7(3): 317–343. Scheltjens, W. 2015. Dutch Deltas: Emergence, Functions and Structure of the Low Countries’ Maritime Transport System, ca. 1300-1850. Leiden: Brill. Veblen, T. 1909. “The limitations of marginal utility”. The Journal of Political Economy 17(9): 620–636. Wäckerle, M. 2013. “On the bottom-up foundations of the banking-macro nexus”. Economics: The Open-Access, Open-Assessment E-Journal 7(2013-40): 1–45. Weber, M. 1980 [1922]. Wirtschaft und Gesellschaft: Grundriß d. verstehenden Soziologie. Tübingen: Mohr, 5th edition.

389

30 DEMOCRACY AS AN EVOLUTIONARY PROCESS Isabel Almudi and Francisco Fatas-Villafranca

30.1

Introduction

Paraphrasing Winston Churchill, we may accept democracy as the less detrimental political system among those historically experienced by human societies. However, as soon as we wonder about the meaning of democracy, we find at least two alternative perspectives in political economy which are not equivalent. As Schumpeter (1942) argued, we can distinguish at least two different approaches. i The classical 18th-century view of democracy as a method to make global political choices, oriented to achieve the public good through instituted acts of social choice. Freedom and participation have been considered within this approach, but always emphasizing their integration in the study of rational choice, political consensus, and stable arrangements. It is remarkable that, even in this (somehow static) approach, non-trivial problems have been pointed out by Arrow (1951), modern Welfare Economics (Arrow et al., 2011), and Public Choice theory (Buchanan and Tullock, 1962; Mueller, 1993). These problems refer to the lack of neat aggregation procedures satisfying a number of natural conditions, the meaning of “the public good,” and how to materialize this idea. ii The second perspective is the dynamic approach to democracy suggested by Schumpeter (1942). Hence, democracy should be conceived as an instituted social process of innovative political competition, aiming to persuade, represent and maintain the majoritarian favor of heterogeneous citizens. It is through this process of competition for political leadership that rival candidates (citizens, civil platforms, parties) seek to implement (within a constitutional frame) their “good society”. Note the processual, restless, essentially competitive, and diversity-driven nature of this approach. In this chapter, we depart from this Schumpeterian view of democracy and seek to explore some properties of democratic systems drawing on the principles of modern evolutionary economics (Dopfer and Potts, 2008). This evolutionary approach allows us to deal with heterogeneity, bounded-rationality, local vs global interactions, political innovative competition, and endogenous change (Metcalfe, 1998). As we look at democracies through the

390

DOI: 10.4324/9780429398971-33

Democracy as an evolutionary process

evolutionary lens, we detect that these systems may suffer from what we have identified as the dynamic Paradoxes of Democracy. The dynamic connotation is crucial and relies on the idea that democracy, far from being a harmonious method to distill the public will in acts of choice (elections, parliamentary and committee votes, constitutional agreements), can be conceived as a multilevel process of political competition, in which citizens and rival platforms incessantly struggle for implementing in a peaceful way their envisioned “good society”. We organize the chapter as follows: in Section 30.2, we identify and explain the dynamic paradoxes of democracy. In Section 30.3, we illustrate how this frame can be made operational for theoretical and empirical analysis from an evolutionary perspective. Finally, we argue that the analysis of the dynamic paradoxes can be especially important in contemporary societies, where innovation, fast economic change, and the connections between rising living standards, ethical confrontations, and the lack of consensus on crucial policies are always present.

30.2

Democracy and its paradoxes

The main statement in this section is that democracy, analyzed as a restless competitive process, instituted in frames which support freedom, legal equality, representation, and participation, necessarily faces three dilemmas (Dahl, 1989). These dilemmas, paradoxes (or tensioned contrary forces) are intrinsic to the dialectic operational mechanisms driving real democracies. This realistic approach to democracy, which is consistent with the claim for empirical relevance in evolutionary economics, disregards the contractarian approaches to original positions, reliance on the veil of ignorance, or ex-ante agreements on political principles (Rawls, 1971).

30.2.1

The representativeness vs governance paradox

If we seriously consider the heterogeneous, conflictive, and ever-changing character of capitalist-democratic societies and also bring out their operational challenges, we see that there is rarely a consensus on how to fix complex problems (Dahl, 1989). Diverse values, contrary interests, and incompatible means to achieve goals, not to say political, technical, and financial competition are intrinsic to democratic societies. Hence, the democratic institutions have to deal with the pragmatics of governance while they represent evolving diversity (Novak, 2018). There is no agreement in modern political science as to how to manage the tension between representation in turbulent societies, and effective governance. The methods based on majorities are a solution, but they are not satisfactory because they sacrifice some people’s will in pursuit of functioning. This is the first dynamic paradox of democracy: representativeness vs governance. This paradox has implications for the structures of democratic assemblies and rules of interaction among the representatives. Those optimistic about deliberative processes (Habermas, 1986) argue in favor of open representativeness. Others (Dahl, 1989) are more skeptical and find dubious the statement that opinions tend to meet, are always innocuous, and truth emerges. Democratic societies seek to design constitutional frames and rules that enhance representativeness (avoiding exclusion), but this often entails incorporating (more or less subtle) forms of potential instability that jeopardize governance (blocking coalitions). This paradox is relevant when heterogeneity in ideologically driven (and imperfectly envisioned) political visions (utopias) is high. In this case, the obstacles cannot be removed 391

Isabel Almudi and Francisco Fatas-Villafranca

behind the veil of ignorance so that history, culture, and vested interests are filtered out (Rawls, 1971). On the other hand, those systems oriented to governance may exclude minorities, and perhaps dubiously constitutional ideas; then, they can produce tensions.

30.2.2

The expertise vs direct participation paradox

Likewise, it is well known that ever since the outset of modern theories of democracy, scholars and politicians (Madison, Alexander Hamilton, or Tocqueville) have shown doubts regarding the capacity of citizens to achieve competent positions in policy issues. Thus, Popper (1945) recognizes that in technically-sophisticated societies, the scattered and specialized nature of knowledge limits generalized competence. This is also well recognized by Hayek (1960) and evolutionary economists (Witt, 2003; Metcalfe, 2010). On the other hand, we find powerful arguments for direct participation leading to successful deliberations and collective learning (Habermas, 1986). Nowadays, the dilemma on the role of open citizen participation (ongoing access to social networks and online opinion arenas) in complicated ethical and practical challenges, vs the prevalence of narrowly commissioned prudential experts, is not solved. This is the second dynamic paradox of democracy: the dilemma between expertise (Martin and Scott, 2000), vs open public participation in complex societies (Wilson and Kirman, 2016). It is clear that fixing political problems requires expertise, narrowly focused experience, and specific training. This implies reliance on teams of technicians, scientists, and lawyers. Often, this option may limit the wide participation ideal. The alternative is accepting ongoing participation, superficial knowledge bases, and communicative action between heterogeneous and boundedly rational citizens in endogenously changing settings (Simon, 1957). There have always existed discrepancies between those arguing in favor of citizen judicious opinions (Dewey, 1927), and those questioning the efficiency with which citizens could form an opinion of matters remote from their lives (Lippmann, 1922). But in our times, and mostly considering the confusing information and communication (ICTs) environments, political theorists agree in that citizen’s political knowledge is superficial. This paradoxical situation may be solved through specific institutional structures oriented towards highly informed (perhaps hermetic) groups of delegates; on the contrary, democracies may opt for open participation, at the risk of weakly supported decision-making.

30.2.3 Novelty vs retained practice paradox Finally, both Popper and Hayek agreed in that a constant enemy of open societies was the propensity to “tribalism”. As Popper (1945) states, our societies have evolved as relatively small and simple groups competing with each other. Thus, when citizens feel insecure (this is common in highly innovative, ethically diverse, competitive capitalist-democratic societies), the idealized remembrances of tribalism can re-assert themselves. This opinion is shared by Hayek (1960), although the Hayekian evolutionary arguments on moral, legal, and technological change recommend taking traditional useful practices seriously. From a different angle, Bauman (2000) conception of the liquid modernity as a type of society in which lasting things and inertias give way to the transient, is also useful to tackle the novelty vs retained practice paradox. Bauman points out that the constant relentless change characterizing capitalist-democratic societies often needs points of reference. The lack of references destroys the social ethos. Even accepting that fast innovation has driven prosperity 392

Democracy as an evolutionary process

during the last two centuries (Nelson, 2018), it seems clear that novelties, new problems, and political (even cultural) adaptations are challenging. Often, and for good, new proposals include institutional changes (Dopfer, 2005; Urmetzer et al., 2018). But the dilemma of novelty vs well-established practice (Muñoz et al., 2011; Pyka, 2017) is always there. Evolutionary economics emphasizes that modern societies continuously change from within and evolve. They adapt continuously, modifying technologies and institutional structures (Metcalfe, 2010; Hodgson, 2015). But this adaptation has to maintain sociopolitical stability, representation, and instituted references. Otherwise, tensions and unevenly distributed changes may lead to dangerous rips (Almudi and Fatas-Villafranca, 2018, 2021). Here we see the intrinsic operation of the three paradoxes in any evolutionary economic approach to democracy. It is clear the need of assimilating qualitative innovative transformations, and this implies a good deal of governance and expertise. But sophisticated change also requires participatory representation of heterogeneous and conflicting agents, prudential judgements facing uncertainty, and the consideration of inertias and collateral effects in the passing of new laws or the regulation of novel markets and political arenas. The complete discussion of these paradoxes exceeds the scope of this chapter, but we devote the next section to exploring several issues related to them. And we do it inspired by Mises (1949) in that nothing prevails for long in society if carried out against the dynamics of public opinion.

30.3

Democracy as an evolutionary process: Utopia competition

We hereby introduce an evolutionary formulation to explore certain aspects of the aforementioned political problems. It integrates the political side of economic agents as competitive citizens (or groups of citizens) who try to bring about their conception of “the” good society. They try to do it by supporting specific social utopias (imperfectly defined views of socio-political organization, which are incompatible among themselves but must be handled). The resulting political process will develop in a dilemmatic space defined by the three paradoxes. For the sake of analytical tractability, we begin by considering that in contemporary democracies there is a finite set of utopias/subsytems that compete for social prominence (Dahl,1956) on political positioning when several issues enter together as a compact utopia or subsystem of ideas and actions.Drawing on Montgomery and Chirot (2015) and Almudi et al. (2017), we state that these social utopias may be the following: Five utopias/subsystems = {C, M, V, S, E}: (C) cultural-traditional subsystem (prevalence of cultural/traditional values) (M) free-market subsystem (liberal or neoliberal open market-driven values) (V) civil society subsystem (politically innovative self-managed values) (S) state subsystem (prevalence of central, communal governance values) (E) environmental subsystem (conservationist values on nature). We assume that a society is composed of heterogeneous citizens who are boundedly rational economic agents. Likewise, the whole society can be de-composed into the subsystems of citizens. As above, each subsystem represents a utopia – i.e. each subsystem can be associated with an ideal (envisioned) state-of-the-world characterized by the prominence of a specific (favored) subsystem over the others. An economic agent – citizen – is characterized in our model by her degree of citizenship when promoting a specific utopia. We represent this degree 393

Isabel Almudi and Francisco Fatas-Villafranca

of citizenship or participation by the proportion of his total amount of resources (money, time, ideas) devoted to fostering his vision at t. We shall consider that citizens may position themselves in low (x1), medium (x2 ), or high (x3 ) levels of contribution, 0 < x1 < x2 < x3 < 1. We pose that 0 < x1 < … xi (xi = x1 + a (i 1)) < 1, a > 0, and that (x1, x2, x3) are identical for all subsystems. The total population of citizens in a subsystem shall be distributed among these alternative behavioral patterns at any time. For each subsystem at t let sjt be the share of citizens within subsystem whose level of citizenship is xj . Therefore, 0 sjt 1, and j sjt = 1.

30.3.1

Citizen payoff

We include gains and costs in what we call each citizen’s payoff. This payoff depends on (Denzau and North, 2000; Kahneman, 2003): i the level of individual citizenship (contribution), which is a good for the citizen although it bears opportunity costs; ii the relative size of the citizens’ favored subsystem; iii a double-externality effect through which citizens assess their (satisfactory but costly) level of effort, with respect to that of their subsystem peers. Regarding (i), we assume that the level of participation and commitment in pursuit of a utopia is a source of satisfaction for citizens. But, it implies opportunity costs. With respect to (ii), it is clear that agents devote their resources to implement their utopia – building up new organizations and institutions, shaping other citizens. We pose that when the favored utopia increases its presence in society (gains supporters) this is a source of satisfaction and material benefit for the agent. For (iii), we consider the opportunity cost of citizenship in payoffs, and the perception of the preferred utopia as being in danger because of low commitment of peers. Citizens try to avoid peers’ free-riding, while they get satisfaction from more-committed peers. A way to capture local effects is by the double-externality in (30.1). We use to regulate the intensity of externalities. We represent (i) to (iii) in the payoffs (30.1):

u1t =

t

(1

) + s2t x1

u 2t =

t

(1

)+

(s3t

u3 t =

t

(1

)

s2t x3

s1t ) x2

(30.1)

1, < 1. We define the , 0 t t = 1, 0 < t is the share of subsystem average level of citizenship in, xt = j sjt xj . The average payoff in subsystems at t is ut = j sjt ujt . The social level of citizenship xt = t xt .

30.3.2

Intra-subsystemic evolution

Citizens in our model can endogenously change, both, their level of citizenship in pursuit of their utopia, and they can also change their minds and direction of action by choosing a different utopia. We assume that heterogeneous boundedly rational citizens coexist within

394

Democracy as an evolutionary process

each subsystem. Thus, citizens differ in their commitment levels and they get a specific payoff attached to this contribution level. Since they feel the gains and costs associated with their behavior, they may revise their level of participation. Firstly, when their current payoff turns out to be too high (or too low) as compared to alternative behaviors, the citizens may update their levels of participation. Secondly, we consider a mutation mechanism capturing innovative opinion changes. Thus, we have collective intra-subsystem dynamics through which citizens adapt their contributions depending on their relative payoffs (Appendix, A1). We obtain five intrasubsystem dynamics for the five utopias, driven by systems of differential equations (Appendix, A1) – the payoff function is 30.1. The dynamics are driven by (30.2) and (30.1):

s jt = (1

) sjt ujt

ut

+

1 3

sjt

j,

.

(30.2)

We close the model by incorporating (30.1) and (30.2) in a co-evolutionary process in 3.3.

30.3.3

Inter-subsystem dynamics and co-evolution

Now, it seems reasonable to suppose that those subsystems that engender stronger levels of citizenship and participation will end up gaining a relative presence in society. This effect will take place as long as citizens change their minds and change their utopias and actions because of the influence of the relative frequency and visibility of other citizens’ opinions, the consequential shaping of related institutions, regulations, and overall society, the media, and social networks. In turn, the emergent uneven prevalence of the alternative utopias will enforce at a higher or lower level the payoff of related individual citizens. We assume that those citizens perceiving the relative success of their favored utopia will feel reinforced in their behaviors, material status, and ideas. On the other hand, citizens that perceive how their utopias lose prevalence may change their minds. Moreover, we may also consider the possibility of citizens changing opinions because of novelties. Thus, we can close our model by proposing a (replicator with mutations) system of five differential equations, coupled with the dynamics presented above in (30.1) and (30.2) (Hofbauer and Sigmund, 1998). The system that closes the model may be expressed as follows: t

= (1

)

t

(xt

xt ) +

1 5

30.3.4

, xt =

t

j

sjt xj

(30.3)

Emergent properties

The co-evolution engendered by (30.1), (30.2), and (30.3) can be studied. It will reveal how the social presence of each utopia endogenously changes in alternative settings. We can obtain different paths for participation, novelty, representation vs governance, and irregularity in the co-existence and utopian change. Considering the limits of this chapter, we are just going to suggest analytical strategies and a few results. We synthesize the results in three propositions in which increasingly generalized versions of the model are analyzed. We begin with a result that is valid for eq. (30.3) and (30.2) separately, which holds for any n (positive integer, finite, thus generalizing the cases n = 5 and n = 3). 395

Isabel Almudi and Francisco Fatas-Villafranca PROPOSITION 1:

From equations (30.3), with constant levels of citizen participation and finite n, we obtain that novelty fuels the persistence of diversity. This may erode governance but assures high levels of representation.

Proof.From (30.3), For any i, j, (i i t i t

t t

j ), we have

j j

)(x i

= (1

x j) +

n

1

1

i t

t

j

The unit simplex is forward-invariant for (30.3). Then, this expression shows that if = 0, just the political competition process operates, and the system tends to select the utopia with maximum participation and commitment from its citizens. It is a situation of maximum 0) then the term governance (conformity). On the contrary, if we introduce novelty ( n

1

1

i t

t

j

plays its role. This means, on the one hand, that the set of resting points is now a

subset of that for = 0. Conformity is no longer accessible from the interior of the simplex, and the prevailing states are interior (persistence of diversity). Although the competitive )(x i x j ) tends to rule out utopias with low participation (think of xi > x j ), as j part (1 tends to disappear, the probability of new ideas revitalizing this option grows very strongly through

1 t

j

. On the other hand, if we look at the time evolution of average participation

xt , we see that for the case with novelty, x=0

V (x ) =

1

(x

x¯)

where x is the weighted average, and x¯ the arithmetic mean (which differ in uneven distributions of participation). V (x ) is the variance. Thus, it does not hold the selection of the fittest result for which x = 0 V (x ) = 0 . PROPOSITION 2:

consider that

t

Let us focus now in the joint operation of equations (30.1) and (30.2), and is constant for the time being and = 0. Then, the intra-subsystem dynamics

for each utopia with endogenously evolving citizen payoffs depends on

(1

)

t

. If this ratio is

low, then we obtain intra-utopia conformity on high participation levels. This fosters the prevalence of said utopia in society. If the ratio is high, participation fluctuates and the corresponding utopia may lack presence or even tend to disappear if the phase of low participation coincides with high participation in competing utopias. Proof.See Appendix (A2). This proposition highlights that citizen permeability to peers’ behavior maintains representativeness. However, in what we call Appendix (A2) of the dynamic diversity regime, citizen learning and adaptation may lead to intense fluctuations in participation (often strong decays in participation followed by sudden increases). The

396

Democracy as an evolutionary process

reason is the double effect in (30.1) of avoiding free-riding vs the need to suddenly enter to support your vision when you perceive it in danger of disappearance (which may be too late). These effects may instill confrontation in society which may erode governance. Note that if we consider together (30.1), (30.2), and (30.3), with co-evolution and time-variant t in inter-utopia competition, the role of the parameter becomes crucial. From proposition 1 above, it also follows that the role of novelty reinforces the complexity arising from fluctuating participation, since it strengthens the prevalence of interior states in (30.3) and rules out high values of t . PROPOSITION 3:

We consider equations (30.1), (30.2), and (30.3) in co-evolution, for any “n 5”. This is a very general case with maximum level of representativeness within utopias and inter-utopias. The dynamics of political competition and participation depend on the values of the intra-utopian ratios (1 ) with respect to the vector of participation degrees t

( xa1 , xa2 , …, xan ) (as abundant and almost-continuous as we want) within each subsystem. The possibility of chaotic unpredictable political dynamics and bifurcations depending on

(1

)

t

appear. Proof.In this case of maximum level of representation, the dynamics of participation and political competition (with and even without novelty) are extremely complex. The challenging proof leads us to pose here just a discussion of the proof (Fatas-Villafranca et al., 2011). In this case, it is remarkable that the political dynamics may even show chaotic fully unpredictable behavior. Attracting heteroclinic cycles appear in the boundary of intra-utopian simplexes thus leading to chaotic behavior for (1 ) sufficiently high. Citizen participation fluctuates t

erratically in the chaotic regime, high levels of representation persist and the case with random novelty intensifies complexity. Governability and expertise-driven behavior in this regime become problematic. The stabilization of retained practices is not assured. Nevertheless, depending on initial conditions, there are settings in which the system traverses for paths close to the simplexes’ edges, thus engendering (as an emergent property) two prevailing options. Still, the paths eventually change cycle and lead to turbulent patterns.

30.4 Concluding remarks In this chapter, we have drawn upon modern evolutionary economics to delineate an evolutionary approach to democracy. This approach has revealed dilemmas that democracies can manage, but they must be explored in a dynamic way. The chapter shows that this is a fruitful approach that can add fresh insights to those obtained by Public Choice theory and welfare economics. Our approach relies on citizens (and groups of citizens) as agents of political change, which competitively shape their socio-economic, cultural, political, and environmental context. This framework may become significant in digitalized societies, characterized by massive individual and group connectivity and mass media. We believe that in this new context, the dynamic paradoxes of democracy may become more relevant than ever before.

397

Isabel Almudi and Francisco Fatas-Villafranca

Acknowledgments This research has been funded by projects PID2019-106822RB-I00,S40 _20R and S40_23R

References Almudi, I., Fatas-Villafranca, F., Izquierdo, L.R. and Potts, J. (2017). The economics of utopia: a coevolutionary model of ideas, citizenship and socio-political change. Journal of Evolutionary Economics, 27, 629–662. Almudi, I. and Fatas-Villafranca, F. (2018). Promotion and Co-evolutionary Dynamics in Contemporary Capitalism. Journal of Economic Issues, 52 (1), 80–102. Almudi, I. and Fatas-Villafranca, F. (2021). Co-evolution in Economic Systems. Third issue in the Series Elements in Evolutionary Economics J. Foster and J. Potts (eds). Cambridge University Press. New York. Arrow, K.J. (1951). Social Choice and Individual Values. John Wiley & Sons. New York. Arrow, K.J., Sen, A., Suzumura, K. (eds). (2011). Handbook of Social Choice and Welfare. Elsevier North-Holland. Amsterdam. Bauman, Z. (2000). Liquid Modernity. Blackwell. Cambridge. Buchanan, J. and Tullock, G. (1962). The Calculus of Consent. University of Michigan Press. Ann Arbor. Dahl, R. (1956). A Preface to Democratic Theory. University of Chicago Press. Chicago. Dahl, R. (1989). Democracy and Its Critics. Yale University Press. New Haven. Denzau, A.T. and North, D.C. (2000). Shared mental models: Ideologies and Institutions. In Elements of Reason. A. Lupia, M. McCubbins and S.L. Popkin (eds). Cambridge University Press. New York. Dewey, J. (1927). The Public and its Problems. Swallow Press. Athens. Dopfer, K. (ed). (2005) The Evolutionary Foundations of Economics. Cambridge University Press. Cambridge. Dopfer, K. and Potts, J. (2008). The General Theory of Economic Evolution. Routledge. London. Fatas-Villafranca, F., Saura, D. and Vazquez, F.J. (2007). Emulation, prevention and social interaction in consumption dynamics. Metroeconomica, 58(4), 582–608. Fatas-Villafranca, F., Saura, D. and Vazquez, F.J. (2009). Diversity, persistence and chaos in consumption patterns. Journal of Bioeconomics, 11, 43–63. Fatas-Villafranca, F., Saura, D. and Vazquez, F.J. (2011). A dynamic model of public opinion formation. Journal of Public Economic Theory, 13, 417–441. Habermas, J. (1986). The Theory of Communicative Action. Polity Press. New York. Hayek, F.A. (1960). The Constitution of Liberty. University of Chicago Press. Chicago. Hodgson, G.M. (2015). Conceptualizing Capitalism. University of Chicago Press. Chicago. Hofbauer, J. and Sigmund, K. (1998). Evolutionary Games and Population Dynamics. Cambridge University Press. Cambridge. Kahneman, D. (2003). Maps of bounded rationality: Psychology for behavioral economics. American Economic Review, 93, 1449–1475. Lippmann, W. (1922). Public Opinion. Simon and Schuster. New York. Martin, S. and Scott, J. (2000). The nature of innovation market failure and the design of public support for private innovation. Research Policy, 29(4), 437–447. Metcalfe, J.S. (1998). Evolutionary Economics and Creative Destruction. Routledge. London. Metcalfe, J.S. (2010). Technology and economic theory. Cambridge Journal of Economics, 34(1), 153–171. Mises, L. (1949). Human Action. Yale University Press. New Haven. Montgomery, S. and Chirot, D. (2015). The Shape of the New. Princeton University Press. Princeton. Mueller, D.C. (1993). The Public Choice Approach to Politics. Edward Elgar. Cheltenham. Muñoz, F.F., Encinar, M.I., Cañibano, C. (2011). On the role of intentionality in evolutionary economic change. Structural Change and Economic Dynamics, 22(3), 193–203. Nelson, R.R. (2018). Modern Evolutionary Economics. An Overview. Cambridge University Press. New York. Novak, M. (2018). Inequality: An Entangled Political Economy Perspective. Palgrave Macmillan. London. Popper, K.R. (1945). The Open Society and Its Enemies. Routledge. London.

398

Democracy as an evolutionary process Pyka, A. (2017). Dedicated innovation systems to support the transformation towards sustainability. Journal of Open Innovation: Technology, Market, Complexity, 3, 27. Rawls, J. (1971). A Theory of Justice. Harvard University Press. Cambridge. Schumpeter, J. (1942). Capitalism, Socialism and Democracy. Harper and Row. New York. Simon, H. (1957). Models of Man. Wiley. New York. Urmetzer, S., Schlaile, M., Bogner, K., Muller M., Pyka A. (2018). Exploring the dedicated knowledge base of a transformation towards a sustainable bioeconomy. Sustainability, 10(6), 1694. Wilson, D.S. and Kirman, A. (eds.). (2016) Complexity and Evolution. The MIT Press. Cambridge. Witt, U. (2003). The Evolving Economy. Edward Elgar. Cheltenham.

Appendix Micro-foundations for the Replicator dynamics and formal analysis

Micro-foundations Let us denote by fij the rate at which citizens contributing xj switch to behavior xi , in their pursuing of more satisfactory behavioral patterns. Let us consider that this switching rate is: fij = [ui

uj ]+ = max ui

uj ; 0 ,

>0

where > 0 captures the ease with which citizens may change their behavior. We are assuming that, given the payoff criteria in (30.1) (Section 30.3), when a citizen from behavioral group i meets another from j, he/she discovers the possibility of adopting behavior xj . Then, we propose that there exists a certain flow of citizens gradually moving in the “better-valuation” direction. Assuming that the product si sj (0 < δ < 1) gives the probability for a random and independent interaction between one citizen with contribution i (share in the population si ) and another one with behavior j (share sj ), in a small interval t, the flow of citizens from j to i would be given by:

si sj fij t and the change in the proportion of citizens with behavior xi would be: si =

j

si sj fij

f ji

t,

fij

f ji = (ui

uj ).

Additionally, we consider that the agents may also choose their contribution randomly with a small probability , due to novelty factors. If we add this component to the replicator equation above, we obtain the following:

si = (1

)

j

si sj (ui

uj ) t +

1 n

si

t

The term that accounts for random experimentation is composed of the outflow si (which is proportional to the number of agents choosing xi ) and the inflow n (which, given that it is

399

Isabel Almudi and Francisco Fatas-Villafranca 1

random decision, and there are n possible alternatives, is proportional to n ). Thus, the net flow (s

s)

of agents changing their contribution from xi to xj because of random experimentation is j n i . Therefore, the continuous time evolution of the proportion of citizens with participation i may be described by the following equation (Fatas-Villafranca et al., 2011):

dsi = (1 dt

si

)

j

= (1

) si

= (1

)

si sj fij s j j

f ji +

( ui

si ui

uj ) +

su j j j

+

1 n 1 n 1 n

si si , si

or, changing velocity (i = 1, .. , n):

si = (1

) si (ui

1 n

u¯) +

si ,

u¯ =

j

sj uj .

In this way, we obtain the expressions in (30.2) (Section 30.3) and – through a similar reasoning – the system (30.3) for the relative share of subsystems in society. If we consider the traditional case in which = 0, then we obtain a typical replicator dynamics system with endogenously changing payoffs:

si = si (ui

u¯),

u¯ =

j

sj uj .

A formal analysis for intra-subsystem dynamics Let us consider the analysis of one subsystem in isolation, with the intra-subsystem structure being composed of three levels of citizen contribution:

x1, x2 = (1 + a ) x1, x3 = (1 + a ) x2,

a > 0.

We consider this representation for mathematical simplicity. In Fatas-Villafranca et al. (2009) we can see that the results are qualitatively similar to the ones obtained for an arithmetic progression among contribution levels of different citizens. Considering the equation (30.1) in Section 30.3, we may re-write them as follows:

ui = xi + (si +1

si 1) xi ,

,

(0, 1),

(i = 1, 2, 3)

If we now combine this simplified expression with the replicator system without mutations

si = si (ui

u¯),

u¯ =

j

sj uj .

we can use both expressions and see that the intra-subsystem dynamics can be analyzed by exploring the following system:

400

Democracy as an evolutionary process

s1 = s1 ( x1 + s2 x1 u¯) s2 = s2 ( x2 + (s3 s1) x2 s3 = s3 ( x3 s2 x3 u¯)

u¯)

(A2.1)

Given that the unit simplex is not altered by the flows induced by (A2.1), the dynamics of (A2.1) are essentially driven by the plane system (s3 = 1 s1 s2 ):

s1 = s1 ( x1 + s2 x1 uˆ) s2 = s2 ( x2 + (1 2s1

s2 ) x2

uˆ = s1 ( x1 + s2 x1) + s2 [ x2 + (1

(A2.2)

uˆ) 2s1

s2 ) x2 ] + (1

s1

s2 )( x3

s2 x3).

In Fatas-Villafranca et al. (2007) we have analyzed a mathematical system which is formally identical to (A2.2). There we proved what in our current case would be the intra-subsystem dynamics driven by (A2.2), depending on specific parametric conditions. More precisely, our intra-subsystem dynamics are determined by the following parametric conditions: a If we set > a , that is to say, when the influence of intra-subsystem local peer interactions is sufficiently intense (beyond a critical level a ), then the interactions between citizens with distinct behavioral patterns will maintain the subsystem in an indefinite process of endogenous self-transformation. There will be an ongoing process of cycling flows of citizens updating their levels of participation even within their current utopia. This is what we called in Fatas-Villafranca et al. (2007) a dynamic diversity D-Regime. b On the contrary, when we have < a , then we obtain a conformity C-Regime in which citizens tend to concentrate gradually on either one or two levels of high participation. This means a concentration of citizen behavioral patterns with time. For the sake of Section 30.3 above, we take into account these results.

401

31 PUBLIC ENTREPRENEURSHIP IN ECONOMIC EVOLUTION Jan Schnellenbach

31.1

Introduction

Public entrepreneurship is not a well-defined concept. There are several different meanings, including entrepreneurship that takes place within the public sector, as well as entrepreneurship in the private sector that is steered or incentivized by public sector decisionmakers. The latter is now in particular associated with the ideas of an entrepreneurial state (Mazzucato 2013), and a mission-oriented innovation policy that assigns a large role to governments in defining where entrepreneurial activity is deemed desirable (Mazzucato 2021). This approach, while immensely popular among political practitioners, is now subject to extensive academic criticism (McCloskey and Mingardi 2020; Wennberg and Sandström 2022). In this contribution, we will however focus on public entrepreneurship as entrepreneurship that occurs with respect to core functions of the public sector. Typically, its aim is to increase the efficiency or quality of public good provision, to find rules that work as solutions to social dilemmas, to solve problems of governance and policy-making, or to change the institutions that govern the process of policy-making itself. The focus here is therefore on a narrower, more traditional understanding of public entrepreneurship. In the following section, we will give a brief overview of the different aspects of public entrepreneurship discussed in the literature. Section 31.3 will discuss the political economy of public entrepreneurship. In Section 31.4, the relationship between public entrepreneurship and economic evolution is discussed on a more general level. Finally, Section 31.5 concludes the discussion.

31.2

Public entrepreneurship: A brief overview

The concept of public entrepreneurship can be traced at least to the doctoral dissertation of Elinor Ostrom (1965), where she analyzes the process of finding a governance solution to a commons problem. This process, she argues, is driven by public entrepreneurs who introduce novel solutions, and who, more importantly, also find ways to implement them by finding incentive structures that make it rational for sufficiently many agents to cooperate.

402

DOI: 10.4324/9780429398971-34

Public entrepreneurship in economic evolution

Ostrom has revisited public entrepreneurship thus understood throughout her career, particularly in empirical studies of communities that solve commons problems by finding effective rules of governance themselves, without the help of centralized social planners with superior knowledge. Public entrepreneurship thus understood is a key element of the Ostrom approach to solving commons problems (most notably Ostrom 1990). It can, however, be generalized to other social dilemma settings, such as the decentralized provision of public goods. The agents who become public entrepreneurs can come from a diversity of backgrounds. Often, it will be professional politicians who think about a reform of given institutions, or about new ways to provide public goods. Sometimes, it can also be bureaucrats who propose means of improving the efficiency of providing public services. In the framework of Ostrom (1965, 1990), public entrepreneurs are often also citizens who do not hold a high-ranking office, but who, as members of their communities, propose rules to govern local or regional commons problems and who attempt to gather political support for their proposals. It is important to note that private and public entrepreneurship are very distinct activities, due to differences in the objects of entrepreneurial activity, and the decision-making environment within which entrepreneurship takes place. Private entrepreneurship with its focus on goods, services, or production technologies, is often highly disruptive because it can be undertaken by individual firms or even individual entrepreneurs. The degree of coordination necessary to initiate private entrepreneurship is often relatively low. In contrast, public entrepreneurship takes place in a political and/or bureaucratic environment, and it depends on the consent of a usually much larger number of veto players than private entrepreneurship. Ostrom and Ostrom (1977) have analyzed the distinguishing features of public entrepreneurship in detail. The first problem is that public entrepreneurship is generally concerned with public goods. This can be true either in the direct sense that a certain public good is provided in some novel form, or in the more indirect sense that there is a reform of policies, regulations, or even decision-making procedures. Even in this indirect sense, initiating a reform, for example, to increase the efficiency of government, is a public good in itself: Initiating a reform benefits an often very large number of individuals, who can however individually free-ride and avoid any individual costs for political support of a reform project. They do not even need to reward a successful policy reform with a vote for the public entrepreneur and probably will not do so if they deem other political issues more salient or personal trust for other political candidates more important. Citizens who are unhappy with the result of public entrepreneurship cannot simply shift their demand to competing suppliers, as they normally can with private goods. The option of individual exit is often prohibitively expensive for public goods, as it requires leaving a community, a state, or a country. Even in systems of competitive federalism, the costs of moving from one subfederal entity to another are often so high that citizens will not move even if public goods provided elsewhere are more attractive. Another issue raised by Ostrom and Ostrom (1977) is measurement. The performance in supplying public goods, as well as the impact of a policy reform, often becomes visible only after a long delay. And if an impact is there, the exact causal influence of entrepreneurial activity in the public sector on an outcome variable is often difficult to disentangle from the impact of other events or policy measures that have occurred during the same periods of time. These issues are technically difficult even for empirical economists, and often contested among them. Even in cases where the science is relatively settled, we know from 403

Jan Schnellenbach

practical examples such as reforms of minimum wage legislation that the consensus evaluation among economists may not become dominant in policy-making circles, let alone among the general public. Boyer and Bang Petersen (2018) present a framework to explain why widespread beliefs about the economy often deviate from the scientific perspective. Therefore, the feedback mechanisms (both positive and negative) for politicians as public entrepreneurs are much more indirect than they are for private entrepreneurs on markets. The same holds for public entrepreneurs as bureaucrats, where rewards for success and punishments for failure are usually much less pronounced than in the private sector. Public entrepreneurship therefore often must be driven by other motives than pure material rewards. This is particularly important because the activity of public entrepreneurship is a voluntary contribution to a non-excludable public good, namely the implementation of a novel policy that solves some societal problem from whose solution also those individuals benefit who have not contributed anything in the process. And it may come at some cost: A politician, a bureaucrat, or a citizen proposing novel solutions risks arguing against the material interests of some groups in society, therefore alienating them and losing their support. There are different candidates in economic theory that can help explain why some individuals are motivated to incur the costs of public entrepreneurship. Career concerns may play a role if individuals expect that successful public entrepreneurship will make them stand out in the competition for higher offices (Kotsogiannis and Schwager 2006). Models of voluntary contributions to public goods (Villeval 2020) do not help very much in explaining motivations of public entrepreneurship, since they generally work well when individual contributions are relatively small and when reciprocity between a larger number of contributors works. They do not help to explain the large voluntary effort behind public entrepreneurship. Not surprisingly, case studies of public entrepreneurship therefore often point out the important personal characteristics of public entrepreneurs. Among them are leadership skills, the skill to identify political windows of opportunity, but also the desire and ability to maximize personal benefits (again, career opportunities are an example) in the process (Cohen 2012; Schneider et al. 1995). Similarities to skills required for private entrepreneurship have been emphasized in several contributions (Klein et al. 2010), and with regard to skills such as alertness to opportunities as discussed in the theory of innovation of Kirzner (1973). But a word of caution seems to be necessary at this point: Some contributions to the literature emphasize the commitment of public entrepreneurs to public or collective values (Waddock and Post 1991). Ostrom herself also focuses primarily on cases where the solution to commons problems yields real welfare improvements. However, public entrepreneurs with idiosyncratic political preferences or with strong personal rather than societal motivations could also use the introduction of novel policies and institutions to their own benefit. This leads us to the political economy perspective on the issue, which will be discussed in the subsequent subsection.

31.3

The political economy of public entrepreneurship

Schumpeter (1942, pp. 290–294) discusses the prerequisites for successful political leadership. With striking similarities to his theory of private-sector entrepreneurs, Schumpeter emphasizes the utmost importance of the personal characteristics of politicians. He demands them to set high moral and ethical standards for themselves and follow the informed recommendations given by experts on specific policy issues. He shows little concern 404

Public entrepreneurship in economic evolution

for institutions such as formal checks and balances that restrict self-interested politicians and rather hopes that a well-functioning selection process will award political offices to an able class of politicians with good motives for pursuing political careers. This view is in sharp contrast to later developments in Public Choice theory, which posits that politicians will be as motivated by self-interest as everyone else, and thus emphasizes the importance of institutions for political outcomes. Here, therefore, the question is which institutions can be expected to spur, and which institutions can be expected to inhibit public entrepreneurship. One important aspect can be the number of veto players that are brought into the process of public entrepreneurship through formal rules (Schnellenbach 2007). A veto player is any individual or organization endowed with the power to block a proposed change of the status quo. For example, the leadership of a coalition party threatening to end a governing coalition can be a veto player. The majority in the second chamber of parliament can also be a veto player. But powerful interest groups, on whose support a government crucially depends, can also act as a veto player, without being formally appointed in a formal constitution. With the number of veto players increasing, public entrepreneurship tends to become more difficult. The argument is similar to the private sector, where we expect organizations with a high degree of bureaucratization and the reliance on a consensus between different departments and individuals to be less agile than organizations that give individual decisionmakers more leeway to act on their own. Even the boldest individual Schumpeterian entrepreneur could achieve little if trapped in a Byzantine system of bureaucracy. But especially from a Public Choice perspective, veto players also play an important productive role in constraining individual politicians in their effort to implement inefficient policies that serve their own self-interest, such as classical rent-seeking policies that restrict productive private-sector entrepreneurship (Baumol 1990). It is therefore not only important how easily a public entrepreneur can implement innovative policies or novel means of supplying public goods. Rather, it is at least as important to make sure that public entrepreneurship is directed in a productive direction rather than aiming at securing rents and appropriating them (Tullock 1989; Choi and Storr 2019). To that end, it is also important to limit government growth, as larger government sizes are empirically associated with lesser entrepreneurial activity in the private sector (Lihn and Bjørnskov 2017). Public entrepreneurship that simply expands the role of government but does not primarily focus on increasing public sector efficiency, is therefore likely to come at the cost of less overall innovation and reduced economic growth. A crucial element to increase both the frequency of public entrepreneurship and its quality may be a decentralization of policy-making. There is a broad literature on federalism as a driver of experimentation with novel policies. This is attractive for two main reasons: Decentralized experimentation with policies is cheaper. If the outcome of an experiment with a new policy is uncertain, and costly failure is a possibility, then smaller-scale innovations in subcentral jurisdictions can produce knowledge about policies without risking large-scale failure (Oates 1999). At the same time, these decentralized political experiments produce positive knowledge externalities that can be used elsewhere. And the fact that voters can monitor the success of policies in their own jurisdictions relative to the success of policies in neighboring jurisdictions also exerts some disciplining pressure on their political representatives (Besley and Case 1995). The vertical dimension of federalism can also provide incentives for sub-central policymakers to conduct risky policy innovations if successful political entrepreneurship increases their chances to win federal offices (Kotsogiannis and Schwager 2006). 405

Jan Schnellenbach

These examples illustrate how formal institutions can incentivize public entrepreneurship, but they also again caution that public entrepreneurship is not normatively desirable per se. It will not necessarily and not always be welfare-enhancing. Another important point that needs to be addressed more explicitly at this stage is the interdependence of both knowledge and subjective beliefs. Ideally, public entrepreneurship generates knowledge about which policies work under which conditions. But as already mentioned above, there are also subjective beliefs of individuals in which they are invested. Economics has been aware of this problem at least since Denzau and North (1994) proposed taking into account what they called shared mental models, and since Caplan (2001) discussed rational irrationality, i.e. the propensity to hold on to political beliefs even though they are not congruent with facts. When individuals are invested in their political beliefs and are reluctant to change them, this implies that we cannot expect the positive knowledge externalities produced by public entrepreneurship to be used efficiently. Substantial opportunities to apply policy-related knowledge, e.g. through imitation of successful policies will remain unused, and it is important to see that this is not only due to standard reasons such as interest group activity (Drazen 1996), but also due to stable political beliefs that are defended against new knowledge that would challenge them. Politicians who know that they are confronted not only with strong interest groups defending the status quo, but also with citizens invested in status quo beliefs will become even more reluctant to become bold public entrepreneurs. A major prediction from a political economy perspective is therefore that public entrepreneurship will occur primarily in times of crisis (Schnellenbach 2007). Seen from this perspective, public entrepreneurship is not to be understood as a careful, deliberate design of policies and formal political institutions by prescient representatives. Rather, public entrepreneurship usually is a reactive activity that occurs once the status quo has become untenable. And even then, it sometimes requires exceptional political personalities with Schumpeterian characteristics to overturn political stasis and implement reform policies, while countries that lack such personalities in office remain in highly inefficient politicaleconomic equilibria for lengthy periods. What this somewhat pessimistic view implies for the interaction of public entrepreneurship and economic evolution will be discussed in the following section.

31.4

Public entrepreneurship and economic evolution

One key insight of the Ostrom approach to understanding the emergence of solutions to social dilemmas is that diversity in institutional arrangements will almost certainly appear (Ostrom 2005). Finding rules that work as solutions to social dilemmas requires coping with societal and economic complexity, and in the absence of a benevolent and omniscient social planner, this calls for experimentation with different rules and different combinations of rules. It is also a process which is highly path dependent. Simply copying a rule that works well in a different polity may not lead to acceptable results, because complementary institutions are missing or exhibit a bad fit to the copied rule. As Ostrom (2005) points out for commons problems, situational characteristics such as the number of agents involved, the level of trust and reciprocity among them, the observability of resource use, and many more influence which rules may be effective and to what degree. This also implies that we are unlikely to observe a complete convergence of rules of governance. This prediction is in line with key arguments made in different strands of literature, 406

Public entrepreneurship in economic evolution

including the emphasis on historical path dependence in the evolution of institutions (in particular North 1990, 1994) and the observation of persistent between-country differences in institutions that coordinate economic agents, as analyzed in particular in the literature on varieties of capitalism (Hall and Soskice 2001). Both approaches emphasize that the prospect for efficiency-enhancing reforms of institutions of policies is often limited. In addition to institutional complementarities, this can also be due to uncertainty about the precise distributional impacts of reforms (Hall and Thelen 2008), and due to the costly nature of individual learning about how to act within a new institutional framework (North 1994). We can therefore expect different speeds of evolution in the public and private spheres, with public entrepreneurship generally moving at a slower pace. This is, however, not necessarily a problem. On the contrary: The formal institutions set by governments are supposed to provide a relatively stable framework that enables market participants to form reliable expectations, coordinate their actions, and realize mutual gains from trade by voluntary agreement. It thus reduces transaction cost. The same holds for public goods, whose predictable supply also contributes to a stable framework that lowers transaction cost for market participants. A highly volatile public sector with a high frequency of creative destruction of institutions or existing public good supplies would call into question the ability of governments to provide stable, predictable environments that reduce transaction costs in the private sector. Taking also the issue of collective conservatism discussed above into account, public entrepreneurship is then often not so much driven by politicians or bureaucrats actively looking for opportunities to improve public sector performance, but by the necessity to respond to crises. Consider a canonical example by Ostrom (1990): When the unregulated commons incentive structure led to an overuse of freshwater basins in California, not even the depletion of the common resource itself was sufficient to initiate successful public entrepreneurship towards implementing effective rules for shared resource use. To achieve this, a ruling by a court was necessary that threatened all parties with undesirable outcomes, so that a negotiated solution that was Pareto-superior to the court’s threat could be reached. This example and the considerations in the preceding section illustrate that it may be overly optimistic to expect public entrepreneurship to play a significant role as an active driving force of economic evolution. This is not so much due to the personal characteristics of politicians and bureaucrats, but due to the incentives that structure their actions. Typically, we can thus expect large-scale public entrepreneurship to occur in response to pressing problems, and only when inaction becomes very expensive or even unsustainable. The expectation that governments should actively choose welfare-improving paths of economic evolution and steer private-sector agents along this path, as it is articulated for example in the mission-oriented approach to innovation policy (Mazzucato 2021), is at odds with the political economy of public entrepreneurship discussed here. To see further how public entrepreneurship can differ between different challenges that need to be met, consider the examples of ozone depletion and climate change. Prima facie both exhibit large similarities as global commons problems. The depletion of the ozone layer due to the emission of chlorofluorocarbons (CFCs) had been discussed as a scientific hypothesis since 1974 and was shortly afterwards supported by first measurements of the ozone layer. It took a mere three years until the Environmental Protection Agency (EPA) in the United States proposed a first ban on using CFSs in aerosols. Federal legislation gave the EPA competencies to take measures to protect the stratosphere in 1977, and a ban on CFCs came into effect in late 1978 (Morrisette 1989). The United States acted as a pioneer, 407

Jan Schnellenbach

joined however by a only few other countries such as Canada and the Nordic countries in Europe. The global coordination problem remained unresolved, and emissions of CFCs increased during the following years. In 1985, a strong sense of urgency appeared when new scientific evidence documented a surprisingly rapid depletion of stratospheric ozone. This sense of urgency relatively quickly led to the negotiation of the Montreal Protocol in 1987, which eventually was ratified by 196 countries and the European Union and led to a quick phase-out of CFCs. Morrisette (1989) identifies three main conditions for the quick negotiating success: i) the clear and consensual scientific evidence and risk assessment; ii) the fact that a negative impact, such as increasing skin cancer cases, was immediate and concerned the broad population; and iii) the fact that the chemical industry had alternatives either already at hand or close to becoming market ready by the mid-1980s. The important third point also was a result of firms increasing their investments in the search for alternatives after the first national-level regulations of the late 1970s. This case shows how pioneering public entrepreneurship (the ban on CFCs in a few countries in the late 1970s) triggered private-sector innovation and how the definitive solution to the problem on a global level was facilitated by these innovations. But it also shows that a solution to the global commons was only possible after cost-efficient technological alternatives existed and after the situation became sufficiently dramatic for incumbent special interests to lose political bargaining power. Morrisette’s second and third conditions are much less clearly established for climate change than they were for ozone depletion. The long-term impacts of climate change are still much less tangible for many individuals, regardless of the existing scientific consensus. Individual effects of climate change are discussed on a more abstract level and are more heavily discounted. Accordingly, democratic pressure on politicians is in many countries still less pronounced than it has been concerning ozone depletion in the 1980s. And the technological advances that allow for a transition to climate neutrality at relatively low economic cost are not ready yet for many applications. We are therefore currently seeing how public entrepreneurship depends on related economic and technological evolution, and also how a failure in public entrepreneurship may be associated with catastrophic repercussions for the further path of economic evolution.

31.5

Summary

Public entrepreneurship is the attempt to introduce new policies or to conduct institutional reforms. Ideally, it occurs with the aim of increasing the efficiency of public good provision, or of introducing solutions to commons problems. It may, however, also be intended to support special interests or to promote rent-seeking. Public entrepreneurship is not always benign. We have also seen that there are good reasons to perceive especially productive, efficiency-enhancing public entrepreneurship as typically reactive. It takes place when the problems that need to be addressed become too pressing to ignore, and sometimes not even then. There is a strong status quo orientation to overcome, and this also requires entrepreneurial personal traits in a politician or a bureaucrat which are not always present when needed. In the field of public entrepreneurship, there are plenty of open questions for further research. Empirical research on the impact of formal and informal institutions on the frequency and scope of public entrepreneurship is still scant. So is micro-level and in particular 408

Public entrepreneurship in economic evolution

experimental research on the personal traits of successful public entrepreneurs. From the perspective of political economy, the question of how to direct public entrepreneurship in a benign direction and avoid special-interest public entrepreneurship is so far debated only to a small extent. Such large gaps in the literature are particularly lamentable because public entrepreneurship will need to play an important role in meeting important challenges like climate change in the near future.

References Baumol, William J. (1990). Entrepreneurship: Productive, Unproductive and Destructive, in: Journal of Political Economy 98(5): 893–921. Besley, Timothy and Anne C. Case (1995). Incumbent Behaviour, Vote-Seeking, Tax-Setting and Yardstick Competition, in: American Economic Review 85(1): 394–414. Boyer, Pascal and Michael Bang Petersen (2018). Folk-Economic Beliefs: An Evolutionary Cognitive Model, in: Behavioral and Brain Sciences 41: 1–65. Caplan, Bryan (2001). Rational Irrationality and the Microfoundations of Political Failure, in: Public Choice 107: 311–331. Choi, Seung Ginny and Virgil Henry Storr (2019). A Culture of Rent Seeking, in: Public Choice 181: 101–126. Cohen, Nissim (2012). Policy Entrepreneurs and the Design of Public Policy: The Case of the National Health Insurance Law in Israel, in: Journal of Social Research and Policy 3(1): 1–21. Denzau, Arthur T. and Douglass C. North (1994). Shared Mental Models: Ideologies and Institutions, in: Kyklos 47: 3–31. Drazen, Alan (1996). The Political Economy of Delayed Reform, in: Policy Reform 1: 25–46. Hall, Peter A. and David Soskice (eds.) (2001). Varieties of Capitalism: The Institutional Foundations of Comparative Advantage. Oxford: Oxford University Press. Hall, Peter A. and Kathleen Thelen (2008). Institutional Change in Varieties of Capitalism, in: SocioEconomic Review 7(1): 7–34. Kirzner, Israel M. (1973). Competition and Entrepreneurship. Chicago: University of Chicago Press. Klein, Peter G., Joseph T. Mahoney, Anita M. McGahan and Christos N. Pitelis (2010). Toward a Theory of Public Entrepreneurship, in: European Management Review 7(1): 1–15. Kotsogiannis, Christos and Robert Schwager (2006). On the Incentives to Experiment in Federations, in: Journal of Urban Economics 60(3): 484–497. Lihn, Jacob and Christian Bjørnskov (2017). Economic Freedom and Veto Players Jointly Affect Entrepreneurship, in: Journal of Entrepreneurship and Public Policy 6(3): 2045–2101. Mazzucato, Mariana (2013). Debunking Public vs. Private Myths in Innovation. London: Anthem Press. Mazzucato, Mariana (2021). Mission Economy. A Moonshot Guide to Changing Capitalism. New York: Harper Business. McCloskey, Deirdre N. and Alberto Mingardi (2020). The Myth of the Entrepreneurial State. London and Great Barrington: Adam Smith Institute and American Institute for Economic Research. Morisette, Peter M. (1989). The Evolution of Policy Responses to Stratospheric Ozone Depletion, in: Natural Resources Journal 29: 793–820. North, Douglass C. (1990). Institutions, Institutional Change and Economic Performance. Cambridge: Cambridge University Press. North, Douglass C. (1994). Economic Performance through Time, in: American Economic Review 84(3): 359–368. Oates, Wallace E. (1999). An Essay on Fiscal Federalism, in: Journal of Economic Literature 37: 1120–1149. Ostrom, Elinor (1965). Public Entrepreneurship: A Case Study in Ground Water Basin Management. PhD dissertation, Los Angeles: UCLA. Ostrom, Elinor (1990). Governing the Commons. The Evolution of Institutions for Collective Action. Cambridge: Cambridge University Press. Ostrom, Elinor (2005). Understanding Institutional Diversity. Princeton: Princeton University Press.

409

Jan Schnellenbach Ostrom, Elinor and Vincent Ostrom (1977). Public Goods and Public Choices, in: Emanuel S. Savas (ed.), Alternatives for Delivering Public Services. Boulder: Westview Press. Schneider, Mark, Paul Teske and Michael Mintrom (1995). Public Entrepreneurs: Agents for Change in American Government. Princeton: Princeton University Press. Schnellenbach, Jan (2007). Public Entrepreneurship and the Economics of Reform, in: Journal of Institutional Economics 3(2): 183–202. Schumpeter, Joseph A. (1942). Capitalism, Socialism and Democracy. New York: Harper. Tullock, G. (1989). The Economics of Special Privilege and Rent-Seeking. Heidelberg: Springer. Villeval, Marie Claire (2020). Public Goods, Norms and Cooperation, in: G.M. Capra et al. (eds.), Handbook of Experimental Game Theory. Cheltenham: Elgar, pp. 184–212. Waddock, Sandra A. and James E. Post (1991). Social Entrepreneurs and Catalytic Change, in: Public Administration Review 51(5): 393–401. Wennberg, Karl and Christian Sandström (eds.) (2022). Questioning the Entrepreneurial State. StatusQuo, Pitfalls, and the Need for Credible Innovation Policy. Cham: Springer.

410

32 EVOLUTIONARY POLITICAL ECONOMY Manuel Scholz-Wäckerle

32.1

Introduction

This chapter aims to present and discuss evolutionary political economy as a research programme, following two central goals. First, to investigate and understand the endogenous dynamics of capitalist development and second, to use formal computational simulation methods, in particular agent-based modelling, synthesizing this knowledge for communication to a larger audience. Traditionally, evolutionary economics has focused on the economic spectrum of political economic evolution, concerned for instance with processes of innovation and imitation in production and consumption. This focus orientates itself theoretically around variational evolution and aims to discuss how novelty originates, adopts and retains in the economy. The approach, however, faces certain difficulties in expressing the transformation from one such a historical period to another – qualitatively different – one. Evolutionary political economy aims to fill this gap by complementing it with a second process, concerned with the development of social choices, following transformational evolution. Social choices involve the political spectrum and build up slowly through mediated communication processes, as democracies provide them. Agent interaction in the political spectrum is qualitatively different from the information exchange between businesses and households in the economy, which bases on price signals. This contrast provides not just two different backgrounds for agent interaction, but also a different tempo of change. Changes in the economic sphere appear rather quickly, due to the variation of new products and production processes on behalf of innovation and imitation activities. Otherwise, social choices develop slowly along the accumulation of real oppositions in the world. Some social choices are of historical magnitude because they structurally affect the entirety of social conditions and thereby enable the potential for qualitative change. The latter provides terrain for transformation into a further stage of political economic evolution, breaking with the previous history. Variational and transformational political economic evolution are interwoven thereby, but evolving with sufficient autarky and at their own tempo. Institutions serve furthermore as carriers for this interplay, through which asynchronous updating occurs between the

DOI: 10.4324/9780429398971-35

411

Manuel Scholz-Wäckerle

politics and the economics. The institutional environment stabilizes the variational dynamics of the economy, allowing it to grow and expand, thereby extending the lifetime of a specific political economic period. However, the economic and political dynamics tend to diverge over the course of a historical period, in such a way, that institutions may not mitigate emerging contradictions and real oppositions anymore, usually resulting into economic crisis and social conflict. Through mediated communication, political subjects recognize these conflicting tendencies – not without struggle – and slowly build up motive forces of transformational evolution, which may eventually result in a disruptive and discrete break of an economic continuity. Political economic evolution unfolds as a stepwise sequence thereby. This stepwise sequence is not only unfolding in time but also in space and in social structures. Geographical places as well as institutions play a central role in synthesizing political economic evolution, leading to uneven developments in the world and in social networks. These two aspects alone make any analytical representation of the interwoven dynamics complicated, but per se not impossible. Models are important in the social sciences because they serve as maps for navigating in those dynamics. Thereby they contribute to the communication processes in the political as well to the information exchange in the economic spectrum. They have a performative role in political economic evolution, giving sufficient reason to advance the modelling apparatus. Models in continuous time, building upon abstraction-based objects, face difficulties in expressing such interwoven dynamics, because space, social structures and time itself cannot be qualitatively differentiated. Allowing a simultaneity of different timings demands a discretization of time, which brings us to simulation approaches. In order to provide models of political economic evolution, one needs to integrate social relations and structural hierarchies on multiple scale to emulate the inherent complexity and modularity of political economic evolution. This aspect turns attention towards disaggregation and leads us to agent-based models, which have the additional advantage of potential inclusion of an explicit spatial topology. The generativist approach of agent-based modelling enables scholars to develop simulations of ascribed political and economic dynamics, but they are not yet very well integrated as an overview of selected publications shall address. A final outlook provides ideas and potential avenues for the further advancement of such a synthetic evolutionary political economy approach.

32.2

Political economy, evolution and transformation

“How can we hope to succeed with the grand plan of taking our evolution in our own hands?” Georgescu-Roegen (1971, p. 355). This question demonstrates, in a realist position, the radical uncertainty scientists are confronted with, when seriously addressing evolution from a social science perspective. The radical uncertainty still holds today, even in light of great successes in the further advancement of this consideration, as we will discuss in the following. Since the discovery of evolutionary theory, it became evident, that if humans are able to understand the endogenous origination of species, they should also be able to understand their own origination, their place in evolution and their own development. The “grand plan of taking our evolution in our own hands” needs to focus on the dynamics of capitalist development, which led to tremendous changes in natural relations (human and non-human). It is a question of economic reproduction, of re-establishing the social ecological conditions of production by providing some stable continuity for development. The current mode of production appears to be instable and needs a critical discussion therefore. 412

Evolutionary political economy

For this very reason, it is crucial to understand how the capitalist mode of production evolves and how it shapes the social ecological conditions for further reproduction. The evolutionary political economy approach presented and discussed here considers this particular evolution as a stepwise, non-teleological sequence of endogenous change.1 We are particularly interested in the development of industrial capitalism and take classical political economy and its critique as a point of departure. Classical political economists recognized that the forces of change – from feudalism to merchant capitalism and thereafter to industrial capitalism – have been of economic as well as of political nature. Marx was the first to understand that this interplay is constituent to economic reproduction (Marx 1992, Part 7), which in its entirety provides the social conditions for a continuity of circulation and accumulation. The problem arising from such an interrelated perspective, which also coins the difference of Marx’s contribution in relation to Smith and Ricardo, is that economic reproduction is dependent on social choices necessary for its recurrent political legitimation. However, these social choices are not necessarily made in harmony, but are more particularly subject to social conflict and struggle arising from the division of labour. The point is that it is not just about conflicting views or different interests of social classes – these are of course always influential – but that the reproduction within capitalist development leads endogenously to institutional contradictions (symbolic) and real oppositions (material).2 Once contradictions are socially exposed from their hedges, they become the motive forces of transformational evolution since they tend to challenge and jeopardize the previous historical legitimation of reproduction. This is a systemic and structural condition in capitalist development carrying a specific evolutionary threat, namely the tendency of authoritarian rule in exploiting an exposed antagonistic social environment by manipulating the formation of social choices in its interest. From a Marxian perspective, political economic evolution is transformational, because societies become the subject of evolution through the formation of social choices, leading to the development of historical stages. It is therefore necessary to include the political process in the analysis, in order to understand how these social choices crystallize. Marx did understand that the social practice of democracy expresses and reflects the power relations in society, but it also provides the means of emancipation and development to overcome those contradictions, in order to provide a stable continuity for reproduction (Scholz-Wäckerle 2016). Schumpeter also shared this idea of a transformational interrelation between the economic and the political spectrum and the role of democracy within it. Schumpeter is very clear in explaining that the circular flow routinizes economic reproduction over time, making it the locus of quantitative growth (Schumpeter 1934, Ch I). This represents a certain continuity that is disrupted only by progressive forces, in his perspective the entrepreneurs. However, one needs to highlight that especially Schumpeter recognized that the emergence of those forces are to be found only in an interrelated political economic perspective, in the realm of social development. “In this social value system all the conditions of life in a country are mirrored, in particular all ‘combinations’ are expressed in it. … For the economic state of a people does not emerge simply from the preceding economic conditions, but only from the preceding total situation” (ibid., 56, 58). Correspondingly, it needs a broader political economic perspective that aims to expose this conflict-laden totality with its inherent contradictions to understand the structural break, potentially leading to qualitative change and transformation. The core element of his further analysis on the dynamics of innovation initialized by structural breaks, still was the Marxian notion of economic reproduction, but of course adjusted to the novel conditions of 413

Manuel Scholz-Wäckerle

his time. The social relations between entrepreneurs and bankers increasingly influenced the development of industrial capitalism, and credibility was substantial for successful entrepreneurship.3 The notion of finance and its role in industrial expansion adds another component to the motive forces of transformational political economic evolution that are still confronting us with contradictions, especially once finance became a speculative end in itself.4 A few decades later, Schumpeter extended his analysis of capitalist development by looking more closely into the market structures that evolved from the great surges of innovation and the corresponding social environment. He found out that the economic process stagnated due to an increasing concentration of capital, leading to monopolistic, at least oligopolistic, competition (Schumpeter 1942). As discussed by Schütz and Rainer (2016, p. 737),5 Schumpeter’s argument is twofold here. First, it is the monopolistic tendency of late capitalism “making stagnation incumbent” by the evolution of a few highpowered corporations in each industry. Second, these industrial movements, accompanied by trustifications and eventual economic stagnation will lead to sharp criticism among intellectual elites and will invoke formation of trade unions and democratic resistance against private property and entrepreneurship.6 Schumpeter’s point mirrors very well the interrelation of economic variational and political transformational evolution that we have previously outlined as the theoretical core of evolutionary political economy with its different dynamics. The social development may disrupt the institutions protecting the past economic order. The reception of Schumpeter’s work in terms of evolutionary thought was, however, a great scientific success and resulted in central themes of what is known as evolutionary economics today. The evolutionary economic dynamics of the 20th century’s industrial capitalism can be very well described along the concept of a Schumpeterian meso trajectory (Dopfer 2012). The idea follows majorly a variational evolutionary perspective and outlines how the origination of an economic novelty (an invention), adopts and retains in the economy as an innovation. The introduction of a meso layer between micro and macro is highly relevant for this undertaking, since it endogenizes the social relations of production and exchange. Meso units, understood as ensembles containing a diversity of subjects (heterogeneous agents such as households, firms, banks) as well as objects (social institutions, commodities, technical artefacts), are considered as the drivers of economic change. An evolutionary economic regime is constituted by a meso trajectory thereby. The question concerning the “grand plan of taking our evolution in our own hands” needs to confront further the mechanisms of a regime transition between two such meso trajectories. This macroevolutionary aspect is not yet concretely worked out, along a transformational interrelation of political and economic dynamics. The proposal of evolutionary political economy focuses therefore on complementing the evolutionary economic analysis by investigating the cumulative formation of social choices in the interrelations between political and economic processes. Correspondingly, such an undertaking implies changing the status of society from an object of (variational) evolution to a subject of (transformational) evolution. This broader perspective is taken from evolutionary biological insights critical to a pure variational view of evolution. As Levins and Lewontin (2009, p. 87) point out: “Darwin’s variational theory is a theory of the organism as the object, not the subject, of evolutionary forces. Variation among organisms arises as a consequence of internal forces that are autonomous and alienated from the organism as a whole. The organism is the object of these internal forces, which operate independently of its functional needs or of its relations 414

Evolutionary political economy

to the outer world.” It is not by coincidence that this Darwinian view has certain logical as well as historical similarities with the ideas of classical political economists, where the participants in the growing capitalist market economy, appear as objects governed by the internal forces of the market mechanism. This was certainly true for the working class, to some extent this might also hold for the mass of businesses in that time. However, society cannot be interpreted as an organism, that is why one needs to introduce a hierarchical, nested or modular approach allowing sufficient autonomy of greater systemic wholes. For the evolutionary political economy, it seems important to refer to a transformational view of evolution, “…, in which the collection of objects evolves because every individual element in this collection undergoes a similar transformation …” (Liagouras 2013, p. 1258 citing Lewontin 2001, p. 53). This perspective reflects very well the idea of the slow transformational evolution of social development, where social choices build up cumulatively. Institutional settings may change abruptly thereby and break with history once “every individual element … undergoes a similar transformation.” Otherwise, Gould (1982) shares a slightly different macrovariational view where the modern synthesis7 “will be supplanted by a hierarchical approach recognizing legitimate Darwinian individuals at several levels of a structural hierarchy, including genes, bodies, demes, species, and clades” (ibid., p. 384). It is modularity and multilevel selection assigning agency to such systemic wholes.8 On behalf of Lewontin’s view, it is further assumed that societies are themselves subjects of transformational evolution. Recently, this idea found also attention in evolutionary economics, in a critique of generalized Darwinism (Liagouras 2013) and a proposal for more structural explanations by incorporating ideas from evolutionary developmental biology (Liagouras 2017). Clearly, scientific developments in evolutionary biology as well as economics indicate that reductionist generalizations are problematic. Synthetic approaches are demanded, since the dialectical de- and reconstruction of theoretical categories and concepts allows to sharpen the view on the particular. In terms of evolutionary political economy, the most obvious synthesis may be given by integrating different themes in heterodox economics, as proposed by O’Hara (2007).9 Indeed, as already highlighted in the introduction, institutions appear as carriers in the interwoven dynamics of political economic evolution. The formation of social choices is governed by the dynamics of institutional change. Democracy appears as a modular institutional structure in this perspective, itself transforming from within a multitude of communication processes, mediating antagonistic values, contradictions and real oppositions (Scholz-Wäckerle 2016). This is a cumulative process affecting also everyday life practices, beyond the micro democratic aspects of voting and elections. Institutionalists have always focused on this interdependency, as demonstrated in the work of Bush (1987). Power and knowledge relations are central for a meso-centred understanding of political economic evolution, where ceremonial values clash with instrumental ones (Veblen 1899). Veblen’s theoretical apparatus highlights how those values are shaped by the economic and social relations, culture, patterns of behaviour and habits of thought. For instance, the conspicuous dynamics of status emulation and imitation in the working class and those of social distinction in the capitalist class, work as stabilizers of aggregate demand, thereby providing continuity for economic reproduction. As shown in the agent-based simulation study by Rengs and Scholz-Wäckerle (2019), once those values are questioned and imitative behaviour of the working class comes to a hold, the economy destabilizes. Such a kind of emerging institutional contradiction may break the variational economic continuity and escalate further societal tensions. Conflicts and eventual institutional collapse may be the outcome, invoking transformational change (Turchin 2003, Elsner 2021). 415

Manuel Scholz-Wäckerle

The Marx-Schumpeter-Veblen complex associates evolutionary transformation with major endogenous changes in the qualities of the social fabric. However, those changes are – as acknowledged by all three seminal scholars – not unfolding in societal harmony, but by social conflict (e.g. class struggle, discrimination of minorities or even armed international conflict). The development of values, knowledge and power relations under antagonistic premises – thereby of hegemony in the social and global division of labour – plays a particular role in the formation of social choices and in evolutionary political economy in general. This evolutionary interrelation of continuous economic variation and disruptive political transformation unfolds as a path-dependent, but irreversible and irrevocable development process (Georgescu-Roegen 1971, pp. 196–200). The arrow of time has just one direction. The idea, that all biological life-forms are subject to a general struggle for entropy,10 is expanded by Georgescu-Roegen (1971, p. 307) to the social realm, “only in the case of the human species has the struggle taken also the form of a social conflict.” This point redirects us to the introductory question about “the grand plan of taking our evolution in our hands”.

32.3

Models, maps and agent-based simulation

Models play a decisive role in the communication processes shaping the cumulative formation of social choices, by influencing values, behaviours and eventually habits of thought. Models are maps (Miller and Page 2007, pp. 36–39) and through this navigational category, they reach performative status in political economic evolution. A realist perspective of science suggests that models should correspond ontologically with the state space they want to emulate. Otherwise, in case of economic models, they may distort expectations leading to collective dissonance. As many have commented, this concerns foremost economic models, building on a general equilibrium framework and a utility-maximizing representative household/firm. But, in general it concerns all forms of reductionist functionalism in the social sciences. A complete retreat to variational evolutionary concepts in political economy, in particular to a selectionist reductionism, would result in similar problems of ontological correspondence, if we take arguments of the previous section serious. The question about methods in and for evolutionary science is a difficult one: “Do we have any reasons for assuming ex ante that there is a category of evolutionary phenomena that follow any analytical trend?”, as pinpointed by Georgescu-Roegen (1971, p. 208). As a tentative response, he suggested following neither a complete historical nor a complete analytical approach, but emphasized to start with history, “the most we can do is to bend our efforts and discover historical trends in spite of the difficulty and uncertainty of the task” (ibid., p. 209). Where he clearly objects the idea of a universal “law of locomotion”, he was indeed attached to systemic approaches for analytical representations, especially concerned with stock-flow, input-output and flow-funds analysis (ibid., pp. 211–275). Central to these analytical themes is the explicit incorporation of time in order to emulate the irreversibility and irrevocability of evolution. Modelling qualitative change involves several difficulties that have forced researchers to its elimination from their models (ibid., p. 216), but introducing a “discrete distinctness” of potential states of the investigated subjects and objects should work as a proxy (ibid.). Then the focus turns to modelling not just the transition of such individual states but the transition on behalf of feedbacks from social interaction and the environment. An evolutionary political economy approach focuses on agents, their social relations and their spatiotemporality in the evolving system. Novelty in terms of qualitative change does not originate 416

Evolutionary political economy

ex nihilo, but builds up slowly by the institutional formation of social choices, of course in frequent interdependency with the variational economic dynamics. “Indeed, the tensions and contradictions of one stage are actually the motive forces of the change to the next stage” (Levins and Lewontin 1985, p. 86). These re-combinatory dynamics assign continuity and stability to an evolutionary political economic regime – such as industrial capitalism – but at the same time establish the conditions for its instability and potential collapse once the transformation of behavioural patterns (discrete distinctness) is not compatible with the institutional structure anymore (Elsner 2021). Computational simulation can contribute in exposing the tensions and contradictions prevalent in a stage of capitalist development. By adding heterogeneity to an agent population, agent-based objects can be simultaneously in different states, in contrast to abstraction-based (aggregated) objects as used in e.g. differential equation systems (Miller and Page 2007, p. 67). Asynchronous agent behaviour in discrete time helps to reveal the motive forces of transformation to a novel stage, if set in relation through social or spatial structures. On behalf of such a dynamic microstructure, meso components may develop on behalf of the institutional relatedness of carried values, understood as networks between agents, objects and structures. At this point, the difference in agents’ discrete distinctness decomposes the whole system into larger components along a “related distinctness”. These distinct meso components generate their own behaviour with sufficient autarky (ScholzWäckerle 2017). If various meso components are set in dynamic relation and correspond to each other, variational economic continuity can be achieved by generating meso trajectories (Dopfer 2012). Meso-centred Schumpeter-Veblen dynamics can be integrated in agentbased macro models thereby, compare Dosi et al. (2010), Ciarli et al. (2010) or Rengs and Scholz-Wäckerle (2019). The transformational evolutionary system behaviour with regard to the institutional formation of social choices is more complicated to model and simulate, since it needs to emulate the societal tensions stemming from clashing social values and institutional carriers. While in a pure variational perspective, the macro layer just operates as a reflecting feedback loop of meso explorations back to micro agents, transformational system behaviour invokes macroevolution, i.e. a metamorphosis of the whole macro structure, changing the entirety of social conditions. For that purpose, it is necessary to introduce different classes of agents, emulating property relations over the means of production as well as the rather slow political process of institutional valuation. First steps in this direction are to be taken on behalf of exposing the slowly developing motive forces of transformation, that are conflicting with the more quickly changing variation of current meso behaviour.11 Agent-based economic simulations with an integrated feedback from finance12 and/or the environment13 contribute to this direction since it reveals contradictions in the current political economic architecture and its social ecological implications. Further steps imply a more thorough representation of institutional diversity and its value-mediating role in the political process of social development. Janssen and Ostrom (2006) have also shown that an agent-based approach to the commons may lead to further insights into those transformational processes.

32.4 Concluding remarks The evolutionary political economy is discussed and presented as a research programme that is emphasizing two deeply interwoven processes, on the one hand the variational economic continuity of growth and on the other hand the disruptive political transformation of 417

Manuel Scholz-Wäckerle

development. Political economic evolution is thereby stressed as an unfolding step-wise sequence of historical developmental stages, marked by finitely stable configurations of social and economic reproduction. The continuity of economic reproduction can only be achieved through the recurrent reestablishment of the social conditions of production, by institutional and political legitimization. The conditions of qualitative change are therefore to be found in the broader realm of social development, where social choices are slowly and cumulatively formed by the institutional mediation of social values. Such large-scale historical changes do not only transform the entirety of social conditions, they may also invoke novel meso trajectories of innovation and imitation activities. However, such transformations are shaped by symbolic contradictions in the communication processes and real oppositions in the material world. Once they are revealed, the social environment is subject to conflict and struggle, thereby escalating motive forces of transformation. This aspect is crucially relevant for the most recent changes in capitalist development, where multiple such contradictions have emerged, e.g. automation replacing labour (origin of surplus value), capital‘s relation to nature (sustainability of human and non-human nature) and unequal exchange in the world economy (exploitation in global value chains). Stability in the current economic continuity is seriously challenged by these contradictions and the question is how the institutional framework mitigates social conflict arising out of those or how it endogenously changes itself on behalf of novel social choices, and novel social values. A crucial objective of evolutionary political economy is to expose those contradictions with theory and concrete computational simulations. Through the establishment of better ontological correspondence, agent-based models may contribute to a better understanding of the interwoven economic and political dynamics, which are running at different tempo. Such integrated political economic evolutionary models still need time to develop, but the current research gap, clearly shows the great potential of this direction. By conceptualizing models as maps, simulations may enhance the mediation of communication processes in the institutional formation of social choices. Thereby they are able to support the exposure of inherent systemic contradictions in capitalist development and to make the complex evolving interrelations between politics and economics more explicit.

Notes 1 Compare Hanappi and Scholz-Wäckerle (2017) for an overview and more extensive discussion. 2 Compare Harvey (2014) for an illustrative summary and discussion of a series of endogenously evolving contradictions. 3 Compare Peneder and Resch (2021) for Schumpeter’s perspective on money and the role of finance in early 20th-century capitalism. 4 Compare Minsky (1986) for the destabilizing role of finance in capitalist development. 5 See also the authors’ ( Schütz and Rainer 2016, 735–737) discussion of Veblen’s perspective on the transformation from capitalism to socialism, pointed out on behalf of the machine process, eventually crowding out the ceremonial values and habits of thought of the business enterprise. The latter contradictory forces are discussed a few paragraphs later here. 6 “[T]hat the actual and prospective performance of the capitalist system is such […] that its very success undermines the social institutions which protect it, and ‘inevitably’ creates conditions in which it will not be able to live and which strongly point to socialism as the heir apparent.” ( Schütz and Rainer 2016, pp. 737–738 citing Schumpeter 1942, pp. 53, 145). 7 Huxley (1942) introduced the terminology of the “modern synthesis” in evolutionary biology and Mayr’s (1942) contribution led to a further hardening of this view. Compare Pigliucci and Müller (2010) for an extended evolutionary synthesis.

418

Evolutionary political economy 8 The commonalities, differences and interrelations between Lewontin’s transformational view and Gould’s macro-variational perspective open up a multitude of research questions and objectives for a further strengthening of evolutionary political economy research, by e.g. relating it to Simon (1962) and setting transformational evolution in context of nearly-decomposable structures in the architecture of complexity. See furthermore Callebaut and Rasskin-Gutman (2005) on the role of modularity in a diversity of scientific fields and Wilson (2016) on complex adaptive systems and multi-level selection. 9 It should be noted that such a general transformational view of political economic evolution is at least implicitly shared by many heterodox scholars, ranging from institutionalists, Neo-Marxists, post-Keynesians, long-wave neo-Schumpeterians, critical economic geographers, international political economists as well as economic sociologists. One way – next to theoretical integration, comparison and/or synthesis – to achieve further synthesis is presented in the next section, along the methodological route of agent-based modelling. 10 As originally conceived by Erwin Schrödinger and Ludwig Boltzmann before him, it is neither a struggle for resources nor for energy, but for negative entropy ( Mayumi 2001, p. 47). 11 See Rengs and Scholz-Wäckerle (2019) for a Veblenian meso foundation with endogenous class dynamics in an agent-based macro model and Gerdes et al. (2022) for a multi-regional, multisector agent-based model capturing abstracted institutional tensions of global political economy, arising out of unequal (economic and ecological) exchange in global value chains. 12 Compare Cincotti et al. (2010), Delli Gatti et al. (2010), Riccetti et al. (2013), Dawid et al. (2014) or Caiani et al. (2016). 13 Compare Safarzynska and van den Bergh (2016), Lamperti et al. (2018) or Rengs et al. (2020).

References Bush P. D. (1987). The theory of institutional change. Journal of Economic Issues 21 (3): 1075–1116. Caiani A., Godin A., Caverzasi E., Gallegati M., Kinsella S. and Stiglitz J. E. (2016). Agent based stock-flow consistent macroeconomics: Towards a benchmark model. Journal of Economic Dynamics & Control 69: 375–408. Ciarli T., Lorentz A., Savona M. and Valente M. (2010). The effect of consumption and production structure on growth and distribution. A micro to macro model. Metroeconomica 61 (1): 180–218. Cincotti S., Raberto M. and Teglio A. (2010). Credit money and macroeconomic instability in the agentbased model and simulator Eurace. Economics: The Open-Access, Open-Assessment E-Journal 4: 2010–2026. Callebaut W. and Rasskin-Gutman D. (eds). Modularity: Understanding the Development and Evolution of Natural Complex Systems, Cambridge, MA/London: MIT Press: 3–28. Dawid H., Harting P., Neugart M. (2014). Economic convergence: Policy implications from a heterogeneous agent model. Journal of Economic Dynamics and Control 44: 54–80. Delli Gatti D., Gaffeo E., Gallegati M. (2010). Complex agent-based macroeconomics: a manifesto for a new paradigm. Journal of Economic Interaction and Coordination 5: 111–135. Dosi G., Fagiolo G. and Roventini A. (2010). Schumpeter meeting Keynes: A policy-friendly model of endogenous growth and business cycles. Journal of Economic Dynamics and Control 34 (9): 1748–1967. Dopfer K. (2012). The origins of meso economics: Schumpeter’s legacy and beyond. Journal of Evolutionary Economics 22(1): 133–160. Elsner W. (2021) Collapse. Institutional decline and breakdown, its endogeneity and its asymmetry vis-á-vis emergence: A theoretical frame. Journal of Economic Issues 55 (1): 79–102. Georgescu-Roegen N. (1971). The Entropy Law and the Economic Process. Cambridge: MA Harvard University Press. Gerdes L., Rengs B. and Scholz-Wäckerle M. (2022). Labor and environment in global value chains: An evolutionary policy study with a three-sector and two-region agent-based macroeconomic model. Journal of Evolutionary Economics 32: 123–173. Gould S. J. (1982), Darwinism and the expansion of evolutionary theory. Science 216 (4544): 380–387. Hanappi H. and Scholz-Wäckerle M. (2017). Evolutionary political economy: Content and methods. Forum for Social Economics 50 (2): 157–174.

419

Manuel Scholz-Wäckerle Harvey D. (2014). Seventeen Contradictions and the End of Capitalism. USA: Oxford University Press. Huxley J. (1942). Evolution: The Modern Synthesis. London, UK: Allen & Unwin. Janssen M. and Ostrom E. (2006). Governing social-ecological systems, in K. L. Juddand L. Tesfatsion (eds.), Handbook of Computational Economics II: Agent-Based Computational Economics, Amsterdam: Elsevier, pp. 1465–1509. Lamperti F., Dosi G., Napoletano M., Roventini A., and Sapio A. (2018). Faraway, So Close: Coupled climate and economic dynamics in an agent-based integrated assessment model. Ecological Economics 150: 315–339. Lewontin R. (2001). It Ain’t Necessarily So. London: Granta Books. Levins R. and Lewontin R. (2009) [1985]. The Dialectical Biologist. USA: Aakar Books. Liagouras G. (2013). Lost in translation: Why Generalized darwinism is a misleading strategy for studying socioeconomic evolution. American Journal of Economics and Sociology 72(5): 1255–1286. Liagouras G. (2017). The challenge of Evo-Devo: Implications for evolutionary economics. Journal of Evolutionary Economics 72(5): 1255–1286. Marx K. (1992). Capital. A Critique of Political Economy. Volume I. London and New York: Penguin Books in association with New Left Review. Mayr E. (1942). Systematics and the Origin of Species, from the Viewpoint of a Zoologist. New York: Columbia University Press. Mayumi K. (2001). The Origins of Ecological Economics. The Bioeconomics of Georgescu-Roegen. Routledge: Abingdon. Minsky H. P. (2008) [1986]. Stabilizing an Unstable Economy. New York: McGraw Hill Books. Miller J. H. and Page S. E. (2007). Complex Adaptive Systems. New York: Princeton University Press. O’Hara P. A. (2007). Principles of institutional-evolutionary political economy – Converging themes from the schools of heterodoxy. Journal of Economic Issues 41 (1): 1–42. Peneder M. and Resch A. (2021). Schumpeter’s Venture Money. Oxford: Oxford University Press. Pigliucci M. and Müller G. (2010). Evolution. The Extended Synthesis. Cambridge, MA: MIT Press. Rengs B. and Scholz-Wäckerle M. (2019). Consumption & class in evolutionary macroeconomics. Journal of Evolutionary Economics 29: 229–263. Rengs B., Scholz-Wäckerle M. and van den Bergh J. (2020). Evolutionary macroeconomic assessment of employment and innovation impacts of climate policy packages. Journal of Economic Behavior and Organization 169: 332–368. Riccetti L., Russo A. and Gallegati M. (2013) Unemployment benefits and financial leverage in an agent based macroeconomic model. Economics: The Open-Access, Open-Assessment E-Journal 7 (2013–42): 1–44. Safarzynska K. and van den Bergh J. C. J. M. (2016). Integrated crisis-climate policy: Macroevolutionary modelling of interactions between technology, finance and energy systems. Technological Forecasting and Social Change 114: 119–137. Scholz-Wäckerle M. (2016). Democracy evolving: A diversity of institutional contradictions transforming political economy. Journal of Economic Issues 50 (4): 1003–1026. Scholz-Wäckerle M. (2017). Generating Meso Behavior. In: Routledge Handbook of Behavioral Economics, Roger Frantz, Shu-Heng Chen, Kurt Dopfer, Floris Heukelom and Shabnam Mousavi (eds), 152–167. Abingdon, Oxon; New York, NY: Routledge. Schumpeter J. A. (1934). The Theory of Economic Development. New Brunswick, New Jersey: Transaction Publishers. Schumpeter J. A. (1942). Capitalism, Socialism and Democracy (third ed.). USA: Harper Perennial. Schütz M. and Rainer A. (2016). J.A. Schumpeter and T.B. Veblen on economic evolution: the dichotomy between statics and dynamics. The European Journal of the History of Economic Thought 23(5): 718–742. Simon H. (1962). The architecture of complexity. Proceedings of the American Philosophical Society 106(6): 467–482. Turchin P. (2003) Historical Dynamics: Why States Rise and Fall. Princeton, NJ: Princeton University Press. Veblen Th. (1899) The Theory of the Leisure Class: An Economic Study of Institutions. New York: Macmillan. Wilson D. S. (2016). Two meanings of complex adaptive systems. In Wilson D. S. and Kirman A. (Eds.), Complexity and evolution: Toward a new synthesis for economics. MIT Press: 31–46.

420

33 DIVISION OF LABOR AS CO-EVOLUTIONARY PROCESS OF ECOLOGY, TECHNOLOGY, CULTURE, ORGANIZATION, AND KNOWLEDGE Ping Chen 33.1

Introduction

Modern science started from empirical rather than philosophical issues. Newtonian mechanics originated from studies of planetary motion. Quantum mechanics emerged from radiation and spectral problem. Evolutionary theory was inspired by the ecological issue of resources and population. Classical economics was founded by the observation of the international division of labor (Smith 1776). Division of labor was studied in economics, sociology, and social biology (Smith 1776, Durkheim 1893, 1997, Wilson 1975, Harris 1978, 2006). This chapter will present a new understanding of Adam Smith and the co-evolution of the division of labor with ecology, technology, and culture from a new perspective of complex evolutionary economics. The new science of complexity introduced new ideas for evolutionary economics. There were two misperceptions of chaos and complexity prevailing in pop science. First, equilibrium thermodynamics and linear stability confined within a fragile order which was heat death without structure (Georgescu-Roegen 1971). Nonequilibrium thermodynamics and nonlinear order laid the viable foundation of living order including life cycles and evolving structure (Chen 2005, 2010). Evolutionary history could be described by bifurcation trees in nonlinear dynamics. Complexity economics uncovered close links between Smith, Darwin, and Schumpeter from an evolutionary perspective. New solutions to old problems, such as efficiency, price, money, and debates among Marx, Keynes, Hayek, and Friedman, could be discussed by a unified theory of the general Smith theorem. Section 33.1 is an introduction. Section 33.2 is the true face of Adam Smith. Section 33.3 is the competition and pricing mechanism. Section 33.4 is price, structure, and knowledge. Section 33.5 is the role of money. Section 33.6 is organization and classification of rules. Section 33.7 is culture and institution. Section 33.8 is issues in science philosophy: falsifiability and ongoing evolution. Section 33.9 is from division of labor to co‐ordination of innovation. Section 33.10 conclusion.

DOI: 10.4324/9780429398971-36

421

Ping Chen

33.2

The true face of Adam Smith and the trade-off between stability and complexity

Modern science was born by searching for better solutions to fundamental questions. The origin of the division of labor is the fundamental issue in economics, which is parallel to the issue of the origin of the universe in astrophysics and the origin of life in non-equilibrium physics and evolutionary biology (Alpher & Herman 2001; Prigogine 1996). The evolutionary view originated in economics by Smith and Malthus which inspired works by Darwin, Marx, Veblen, and Schumpeter (Darwin 1859; Marx & Engles 1848; Veblen 1898, 1899; Schumpeter 1939). Neoclassical economists introduced linear math for the steady state of short-term equilibrium in demand and supply. The linear model of a pure exchange economy assumed unlimited resources in production and unlimited capability in consumption induced ecological crises and social conflicts. This chapter will begin with the nonlinear dynamics of the division of labor and the theory of metabolic growth. We discovered the TRUE face of Adam Smith. Complexity economics based on logistic growth (Chen 1987, 2014) was a general form of the Smith Theorem that “division of labor is limited by the market extent” (Smith 1776, Book I, Chapter III; Stigler 1951). In contrast, the Smith Speculation was wrong. Smith believed that international trade balance could be achieved by market forces. The Smith hypothesis was based on simple speculation that Dutch ships that carried grain from Prussia and returned ships that carried fruits and wine from Portugal should have the same value (Smith 1776, Book IV, Chapter II). The Smith Speculation was rejected by the history of persistent trade imbalance and trade wars (Chen 2021). The general equilibrium framework in neoclassical economics broke down under path dependence and increasing return to scale (David 1985; Arthur 1994a). Re-evaluation of the Smith Theorem triggered a chain reaction in neoclassical economics. First, the market extent was mainly determined by the resource capacity. Unlimited growth theories in macroeconomics had a better alternative to the metabolic growth theory (Solow 1970, Romer 1986, Chen 2014). Second, linear utility function and production functions became invalid when nonlinear demand and supply curves were shaped by limited resources. Human nature was not greedy since social animals must cooperate and compete for survival under limited resources. Third, market-share competition is a monopolistic competition based on strategic pricing, the so-called micro foundation of perfect competition and marginal pricing did not play a major role in the industrial economy with increasing returns to scale. The general equilibrium framework breaks down, which is not compatible with increasing returns in production (Arrow and Debreu 1954; Arthur 1994). Technological metabolism mainly occurred at the meso-level of industrial organization. The two-level framework of micromacro in Keynesian economics needs a medium structure as the three-level framework of micro-meso-macro in nonlinear economic dynamics (Chen 2002, 2010). Fourth, the political economy had a better understanding of the market power when the resource competition became the driving force of colonialism and trade war. Smith himself did not explain what wealth was. He simply quoted Hobbs that “wealth is power” (Smith 1776, Book I, Chapter V). The essence of economics is political economy rather than exchange economy in neoclassical economics (Galbraith 1972). Fifth, the ecological network of the global division of labor changed our understanding of economic order from static equilibrium to dynamic order parallel to a biological organism. The issue of complexity and stability of ecological systems sheds new light on the relationship

422

Division of labor as co-evolutionary process

between biodiversity and system stability. We developed a general Smith Theorem or SMC (Smith-May-Chen) Theorem that the international division of labor was limited by scale economy (resource capacity), scope economy (resource diversity), and environmental fluctuations (Chen 2014). The economic evolution of the international division of labor is a two-way co-evolution. Under weak fluctuations, the system would increase network connections and ecological complexity; under strong fluctuations, such as crisis; bifurcation and phase transition would occur, so that complex economic systems would be bifurcated into different paths. Sixth, convergent evolution of optimal approach in equilibrium economics (Alchian 1950) should shift to divergent evolution in evolutionary economics (Veblen 1898). Economic equilibrium was justified by Coase’s argument of zero transaction costs (Coase 1988). Counter evidence was found from the U.S. history that the transaction costs of the U.S. GDP were about 25% in 1870 but 50% in 1970 (Wallis and North 1986). The living structure was created for overcoming frictions under non-equilibrium thermodynamics (Prigogine et al. 1972; Chen 2007) because biological evolution could be explained by the dissipative structure.

33.3

Competition and pricing mechanism: Efficiency, risk, uncertainty, and adaptability

For Adam Smith of classical economics, efficiency in the division of labor implied the increasing return to scale, and the winner controlled the large share of market extent in power competition. Its pricing mechanism was a monopolistic competition by all means. Market expansion was an internal drive beyond the current status. For neoclassical economics, efficiency means status quo since perfect competition leads to zero profit and static equilibrium. The so-called efficient market was a lifeless market that had no creativity and winners at all. For the labor theory of Marx (Marx 1867), calculating average labor time was only possible for mature technology without uneven development. In contrast, the theory of creative destruction by Schumpeter implied great uncertainty caused by business creativity and organizational innovation. Which school was relevant to business practice and macro policy? There are clashing pictures of market dynamics within the community of economics. Recurrent business cycles and financial crises demonstrated that the market was not selfstabilizing. This is the nightmare of the so-called efficient market with perfect information (Fama 1970) and rational expectations (Muth 1961). Information uncertainty was visible in financial crises and trade wars (Knight 1921; Kindleberger 1986). In the real business world, the existence and stability of Fortune 500 companies were in the range of decades. They were all monopolistic companies that controlled leading technology in their sectors. Obviously, the concept of rationality and equilibrium was ambiguous to development theory as a process, not the goal. In thermodynamics, the utopian order in equilibrium economics is the heat death or disorder of isolated and closed systems in equilibrium physics (Georgescu-Roegen 1971). The living orders of biology and society only exist in non-equilibrium thermodynamics in open systems (Prigogine 1980). The question is what kind of non-equilibrium system is relevant to economic phenomena. In complexity economics, there are three competing schools in non-equilibrium physics (Chen 2019). Behavioral psychology observed a small asymmetry between gain and loss aversion which was a small deviation from equilibrium economics (Thaler & Sunstein 2021). Path 423

Ping Chen

dependence and noise trader added historical dimension and heterogeneous behavior disclosed disequilibrium dynamics (David 1985; Shleifer and Summers 1990). On the opposite position, CAS (complex adaptive system) and power law in econophysics belong to a nonequilibrium market on the edge of disorder. For computer simulation of artificial life and the stock market, all sorts of unstable patterns of “emergence” or “self-organization” occurred with little evidence in real economics (Chen 2024). The middle ground was market as a living order from the biophysics perspective. The market resilience could be seen through recurrent crises. This was the basic belief of Hayek and Schumpeter, but not Keynes. The higher order of the “living organism,” first suggested as the dissipative structure by the Brussels school (Nicolis and Prigogine 1977; Prigogine 1978), then observed as color chaos from economic indexes by the Austin school (Chen 1988, 1996a, 1996b), and theorized as the viable market by the China school (Tang and Chen 2014, 2015). Schumpeter’s concept of entrepreneurship developed from individual innovators to visionary governments. The idea of meso foundation would be a two-way thinking from the upward complementarity on creative destruction at the micro players to downward complementarity on crisis management of macro regulators in the financial market (Dopfer 2005, Dopfer et al. 2017; Chen 2005).

33.4

Price, structure, and knowledge

Price theory plays a central role in economics. For quantitative analysis, the basic question is how many variables are needed to characterize price movements. In physics, gravitation force played a central role in Newtonian mechanics. Newton’s gravitation theory was a scalar theory. The distance between two particles is good enough to solve the planet motion problem with one or two bodies. However, the electromagnetic field was a vector theory, and the relativity theory was a tensor theory. The number of variables indicated the degree of physical complexity. The advances of complexity economics discovered increasing complexity in price movements. For the micro theory of classical economics, the price theory was a scalar model of gravitation. The nominal price alone could assure the equilibrium between demand and supply. In the real business world, pricing dynamics in practice were view as a vector model including many variables, such as trading volume, various interest rates, and its sensitivity to political news. Our model of monetary chaos discovered more advanced dynamics in the same category of neural networks (Chen 1988). Hayek considered the price as local knowledge distributed in the whole market that was similar to a big-bang cosmic model (Hayek 1945). The question was how to measure market knowledge based on empirical observations. Hayek’s “self-organization” as “spontanewereous order” did not indicate any computational methods. In business practice, various models of pricing mechanisms can be ranked from simplicity to complexity, which were shaped by varying environments and heterogeneous structures. Anthropologists found the hunter-gather society with only limited wants (Gowdy 1997). The neoclassical model was incapable of understanding a village market with seasonal cycles and climate crises. Industrial development in western Europe was described as a staged nonlinear process; while development in Asia and other continents was better characterized as evolutionary trees (Engels 1884; Rostow 1990). Price dynamics were varied with changing environments and cultural behavior. 424

Division of labor as co-evolutionary process

33.5

Changing nature of money in history, from Smith to MMT

The role of money changed significantly in history (Ferguson 2002; Arrighi 2010; Richards 2012). It started from the exchange medium to debt instrument for war and financial power in national competition. There was a theoretical issue for monetary theory: the nature of money is neutral (Lucas 1972), exogenous (Friedman 1969, 1992; Kelton 2020), or endogenous (Hayek 1975). The discovery of color chaos from monetary indexes gave theoretical evidence of endogenous money (Chen 1988, 1996a). Historical events provided natural experiments to test competing theories of money. Why did the Erhard miracle in West Germany in 1948 succeed but the so-called shock therapy of price liberalization failed in East Europe and the Soviet Union? The answer could be traced to Sraffa (1960). The Erhard Miracle could only occur in an atomic economy when industrial structures and networks were destroyed during World War II. This was not the case for the Soviet Union and Eastern Europe. The industrial structure in the command economy in the Soviet Union and Eastern Europe was vertical integration based on topdown design. Without the commanding system, no unit firm in the product chain could survive in the market system. The consequence of product chain breakdown could be observed in the recent trade war (Chen 2021). The case of the monetary union in Germany on July 1, 1990, provided a historical test of Friedman’s helicopter money (Friedman 1992; Chen 2007). Before currency unification, there were two German currencies: East German Mark (DDR) and West German Mark (DR). Its official exchange rate was 1 to 1. But the shadow exchange rate was about 6 to 1. For political motivation of German unification, the residents in East Germany could covert DDR to DR with a favorable exchange rate under different quotas. This was a real test of the neutrality of “helicopter money” in history. From a monetary perspective, monetary neutrality would predict no real effect on the economy; while Keynesian theory would worry about huge inflation. Unfortunately, complex situations occurred for endogenous money. The real story was deflation and deep recession in the region of East Germany. The monetary transfer from West to East Germany was about half of East German GDP for nearly a decade. Why? Monetarists ignored the critical issue of trade structure. Before the currency unification in 1990, East Germany (DDR) was the most advanced country in the Socialist block. Its export to the socialist countries did not use hard currency in the West. After currency unification, East Germany suddenly lost its big export market because socialist countries could not pay with hard currency. East German industry also lost its significant domestic market when East German residents switch their consumption from cheap products made in East Germany to fancy products made in West Germany with their helicopter money. The result was large-scale bankruptcy and unemployment in East Germany. The stimulating consumption of products made in West Germany was also shortlived. The rising interest rate in DM slowed down the whole German economy and left a hard burden on the future of the European monetary union. Adam Smith acknowledged that “wealth is power” because of the difference between the nominal and the real price of exchanged goods (Smith 1776, Book I, Chapter V.). There is a theoretical issue on the upward and downward complementarity (Dopfer, Potts, and Pyka 2017). It is known that monetary and fiscal policy has different weights in different phases in business cycle policy. In the rising phase of the macroeconomy, business investments were sensitive to interest rate policy. The bottom-top perspective would favor the upward complementarity for monetary policy. However, in a macroeconomic downturn, a large scale

425

Ping Chen

of unemployment could be better managed by the Keynesian fiscal policy, which supports the top-down perspective or downward complementarity. Conflicting philosophical views can be integrated into an endogenous theory of money with life cycles (Chen 2014).

33.6 Evolution of organization and classification of firms Darwin’s theory of species evolution was based on Linnaeus’s classification of plants and animals (Linnaeus 1735; Darwin 1859). Applications of quantum mechanics in chemistry were built on the periodic table (Mendeleev 1869; Pauling 1970). The question is why economics lacks a theory of economic structure or meso foundation, which should have multi-layers in the vertical and networks in the horizontal space. There are several factors that discourage structural theory in economics. First, atomism in classical philosophy favors atomism and the reductionist approach in economics. A notable example was methodological individualism (Weber 1905). Economists had trouble on the concept of social value for non-communistic society (Schumpeter 1909). This difficulty could be solved by contemporary issues like climate change that indicated the social value of the whole earth. The global competition occurred not only at the firm level, but also at the national and regional level. The emergence of the EU and environmental movements was better understood from evolutionary economics with ecological complexity. Second, progress in mathematical tools stimulated new concepts in economics. For example, the representative agent model was designed for static statistics with finite mean and variance. Its implicit assumption was a homogeneous population without conflicting interests between regions, sectors, classes, and nations. The non-stationary and nonGaussian distribution could reveal complex features of a heterogeneous population. The general equilibrium framework treated different types of firms as the same type of economic agents (Arrow and Debreu 1954), which had no room for industrial classification in terms of their size, growth, and technology. Business schools developed several approaches to firm classification for business management, financial investment, and tax policy. For institutional economics, a classification of the R&D organization, budget, and social network was useful for policy evaluation.

33.7

Culture and institutions

We observed that institutions and rules varied greatly across different cultures rooted in environment and history. We are interested in three issues in institutional economics: who and how to make rules and why rules matter. The equilibrium perspective advocated the utopian world of the zero-transaction cost that demands minimum information from the market (Coase 1988); while the Austrian school demanded infinite local information (Hayek 1945, 1979, 1990). The evolutionary perspective proposed a co-evolution of downward and upward complementarity (Dopfer et al., 2017). For Santa Fe school, self-organization implied that even unstable sandpile was possible pattern in economies (Bak et al., 1987). Quantum biology indicated that few innovations could survive because the Principle of Large Numbers, which was valid for the financial market (Schrodinger 1948; Chen 2005; Tang and Chen 2014). From the biology perspective, the function of the institution was similar to the cell membrane that selected nutrition inflow and released waste outflow (Chen 2007). In dealing financial crisis, Keynesian economics 426

Division of labor as co-evolutionary process

was more practical than the Austrian and Schumpeter’s policy in dealing with the Great Depression, since national competition played a major role during the financial crisis. Consider two notable issues of internet technology and the healthcare system for a comparative study of rules on innovation policy and technology regulation. The internet technology was developed in defense projects in United Kingdom, United States, France, and the international physics organization of CERN in the 1950s to the 1970s, and diffused to commercial applications in 1980s (Ryan 2010). China introduced internet only in the 1990s, far behind Japan and Korea. Surprisingly, United States and China developed internet technology ahead of Europe and Japan. The driving force of the internet market was the Defense Department in the United States, but the private sector was supported by local governments in China. Over-regulation in EU and Japan discouraged new technology because of the over-protection of old technology. Certainly, scale economy was crucial in market-share competition for emerging technology. Another interesting issue was the healthcare systems around the world (Commonwealth 2021). If we compare the economic costs and social welfare, Japan and European countries were far better than the United States. The U.S. healthcare systems were so complicated. Its high costs were rooted in the decentralized systems and conflicting interest groups. The increasing complexity of legal rules was driven by lobby groups in the United States. In contrast, China’s legal reform was rapidly simplified through decentralized trial-and-error experiments and coordinated under the consultation system in China. This is why China’s mixed economy is very competitive in global competition. For increasing technology division of labor, the coordination problem plays a more significant role in a sharing economy than the efficiency problem in an exchange economy (Heinrichs 2013; Chen 2021).

33.8

Issues in science philosophy: Falsifiability and ongoing evolution

The development of complexity science and evolutionary economics raised fundamental issues for science philosophy. First, complex systems are hard to test by empirical observation in the sense of falsifiability (Popper 1934). Take the example of BZ reaction in chemistry that was a notable case of nonlinear chemical oscillator and chaos. The simplest BZ reaction model of Brusselator had two equations which included four reactions (Prigogine and Lefever 1968). The more sophisticated Oregonator had three equations that involved 11 reactions and 12 species (Field and Noyes 1974). These chemical reaction models may explain some stylized pattern, but fall short to “falsification” (Popper 1934). Second, both biological evolution and cosmic evolution are ongoing processes without an end or convergent equilibrium (Prigogine 1996). Therefore, both induction and reduction reasoning are open-ended projects for an evolutionary process. Einstein and Popper could argue for potential alternatives with hidden variables in the debate on the nature of quantum mechanics (Popper 1963). Economists could introduce more hidden variables too. Physics variables were expanding from Newton’s mechanics, Einstein’s relativity, to quantum field theory and particle physics. The rise and fall of competing theories were wave-like movements of paradigm shifts from special theory to general theory in a race to explain more empirical facts with a unified theory (Keynes 1936; Chen 2010). Third, studies on brain science and epistemology found a close relationship between mental activity and brain structure (Piaget 1971). The dividing line between the real world and the subjective mind is diluted. Different ecology and technology would deeply influence 427

Ping Chen

diversified civilizations, including their culture, language, thinking way, and behavior. Economics as an empirical science will become more inclusive and diversified.

33.9

From division of labor to co-ordination of innovation

When Smith observed a pin workshop with 10 workers, its output was 48,000 pins per day (Smith 1776, Book I. Chapter I). If one family could buy 100 pins for their needs within a year, this pin workshop needed a market with 200,000 households or a population of 1 million. The market extent of division of labor in production was much larger than Plato’s optimal size of a city-state, which was a pretty number of 5,040 (Plato, Laws 2016). If we consider the family size including slaves, the ideal size of city state without internal divide should be in the order of 100,000, which was one-tenth of the typical city in the 18th century. In the 21st century, Apple company had more than 200 supplier companies across several continents with a market value of more than $2 trillion. In parallel, CERN, the largest research organization in particle physics, had nearly 3,000 staff and more than 10,000 researchers from 70 countries around the world. There were two driving forces in the structural evolution of the international division of labor. One was changes in energy and communication technology which induced changes in power distribution (Toffler 1980). The other is changing geopolitics (Chen 2021). The changing composition of the Dow Jones Industrial Average provided valuable information on the rise and fall of leading industries in the United States (Stillman 1986). The expansion of market extent played a critical role in geopolitics. Historically, the Kondratiev long-wave revealed the bottom-up perspective of technology innovation. The four world systems from Spain, Dutch, British, to American empires made clear that the top-down approach had more weight in the international division of labor (Arrighi 2010). We see that the market extent and market power are increasingly associated with the size of a country or economic union with essential resources and infrastructures.

33.10

Conclusion

The evolutionary perspective introduced history and culture into the path-dependence theory of economic development (Veblen 1898, 1899; Prigogine 1980). Nonlinear dynamics and complexity science developed powerful tools to explore the evolutionary process. For example, genetics and quantum biology greatly improve our knowledge of biological and social evolution (Schrödinger 1948), which were applied in analysis of financial crisis (Tang and Chen 2014, 2015). A new understanding of the Smith Theorem and international division of labor extended our scope of institutions in managing war and conflicts. For future development in evolutionary and institutional economics, mainstream and heterodox economists should work together to advance a paradigm shift, from competing for the wealth of nations to nurturing the coordination of nations (Chen 2022). A new economic science should meet contemporary challenges, such as climate, population crisis, and wars (Chen 2024).

Acknowledgment The author thanks the following for a stimulating discussion: Kurt Dopfer, James Galbraith, Brian Arthur, Gary Jefferson, Wolfram Elsner, Andreas Pyka, and Yi Wen. 428

Division of labor as co-evolutionary process

References Alchian, A.A. (1950) “Uncertainty, Evolution, and Economic Theory,” Journal of Political Economy, 58(3), 211–221. Alpher, R.A., and Herman, R. (1948) “Evolution of the Universe,” Nature, 162(4124), 774–775. Alpher, R.A., and Herman, R. (2001) Genesis of the Big Bang, Oxford University Press, Oxford. Arrighi, G. (2010) The Long Twentieth Century, Money, Power, and the Origin of Our Times, Verso, New York. Arrow, K.J., and Debreu, G. (1954) “Existence of an Equilibrium for a Competitive Economy,” Econometrica., 22 (3), 265–290. Arthur, W.B. (1994a) Increasing Returns and Path Dependence in the Economy, University of Michigan Press, MI: Ann Arbor. Arthur, W.B. (1994b) “Inductive Reasoning and Bounded Rationality: The bar El Farol in Santa Fe,” Amer. Eco. Rev. 84(2), 406–411. Arthur, W.B. (2021) “Foundations of Complexity Economics,” Nature Reviews Physics, 3, 136–145. Bak, P., Yang, C., and Wiesenfeld, K. (1987) “Self-Organized Criticality: An Explanation of the 1/f noise,” Physical Review Letters, 59(4), 381–384. Challet, D., Marsili, M., and Zhang, Y.C. (2005) Minority Games: Interacting Agents in Financial Market, Oxford University Press, Oxford. Chen, P. (1980) “Historical Roots of China’s Science Fall Behind the West”, in Chen, P. ed., Theory of Metabolic Growth: Technology Wavelets, Rise and Fall of Civilizations, Chapter 4, (in Chinese), Peking University Press, Beijing, 2019. Chen, P. (1987) “Origin of the Division of Labor and a Stochastic Mechanism of Differentiation,” European Journal of Operational Research, 30(3), 246–250. In Chen, P. Chapter 8, London: Routledge, 2010. Chen, P. (1988) “Empirical and Theoretical Evidence of Monetary Chaos,” System Dynamics Review, 4(1-2), 81–108. In Chen, P. Chapter 4, London: Routledge, 2010. Chen, P. (1991) “Needham’s Question and China’s Evolution – Cases of Nonequilibrium Social Transition”, in Scott, G. ed., Time, Rhythms and Chaos in the New Dialogue with Nature, Chapter11, pp.177–198, Iowa State University Press, Ames, Iowa (1991). In P. Chen, Chapter 10, pp. 217–230, 2010. Chen, P. (1996a) “A Random Walk or Color Chaos on the StockMarket?-Time-Fre quency Analysis of S&P Indexes,” Studies in Nonlinear Dynamics & Econometrics, 1(2), 87–103. (1996a). In Chen, P. Chapter 6, London: Routledge, 2010. Chen, P. (1996b) “Trends, Shocks, Persistent Cycles in Evolving Economy: Business Cycle Measurement in Time-Frequency Representation,” in W. A. Barnett, A. P. Kirman, and M. Salmon eds., Nonlinear Dynamics and Economics, Chapter 13, pp. 307–331, Cambridge University Press (1996b). In Chen, P. Chapter 7, London: Routledge, 2010. Chen, P. (2005) “Evolutionary Economic Dynamics: Persistent Business Cycles, Disruptive Technology, and the Trade-Off between Stability and Complexity,” in Dopfer, K. ed., The Evolutionary Foundations of Economics, Chapter 15, pp. 472–505, Cambridge University Press, Cambridge. In Chen, P. Chapter 3, London: Routledge, 2010. Chen, P. (2007) “Complexity of Transaction Costs and Evolution of Corporate Governance,” Kyoto Economic Review, 76(2), 139–153. In Chen, P. Chapter 14, London: Routledge, 2010. Chen, P. (2010) Economic Complexity and Equilibrium Illusion: Essays on Market Instability and Macro Vitality, Routledge, London. Chen, P. (2014) “Metabolic Growth Theory: Market-Share Competition, Learning Uncertainty, and Technology Wavelets,” Journal of Evolutionary Economics, 24(2), 239–262. Chen, P. (2016) “Mathematical Representation in Economics and Finance: Philosophical Preference, Mathematical Simplicity, and Empirical Relevance,” in Ippoliti, E. and Chen P. Eds. Methods and Finance, A Unifying View on Finance, Mathematics and Philosophy, pp. 17–49, Springer, Berlin. Chen, P. (2019) “From Complexity Science to Complexity Economics,” in Beker, V. A. Ed. Alternative Approach of Economic Theory, Chapter 2, pp. 19–55. Routledge, London. Chen, P. (2021) “China, and the West, The Metabolic Nature of Changing World Order,” in Svejnar, J. and Lin, J.Y. eds. China and the West: A Commentary on Key Economic Issues, Chapter 9, Edward Elgar.

429

Ping Chen Chen, P. (2022a) “From Economic Chaos to Viable Market,” in Chen, P., Elsner, W., Pyka, A., Eds. Routledge Handbook of Complexity Economics, Routledge (2022, forthcoming). Chen, P. (2022b) “History, Perspectives, and Issues of Economic Complexity,” in Chen, P., Elsner, W., Pyka, A., Eds. Routledge Handbook of Complexity Economics, Routledge (2022, forthcoming). Coase, R.H. (1988) The Firm, the Market, and the Law, University of Chicago Press, Chicago. Commonwealth Fund (2021) Mirror, Mirror 2021: Reflecting Poorly, – Healthcare in the U.S. Compared to Other High-Income Countries, Aug. 2021; https://www.commonwealthfund.org/publications/fundreports/2021/aug/mirror-mirror-2021-reflecting-poorly?gclid=EAIaIQobChMI54rGjvad9wIVDxpMCh 0ILAlqEAAYAyAAEgJT5_D_BwE, Apr.18, 2022. Darwin, C. (1859) On the Origin of Species, by Means of Natural Selection, or the Preservation of Favoured Races in the Struggle for Life (1st ed.), John Murray, London, London. David, P. (1985) “Clio and the Economics of QWERTY,” American Economic Review, 75 (2), 332–337. Dopfer, K. (2005) “Evolutionary Economics: A Theoretical Framework,” in Dopfer, K. ed., The Evolutionary Foundations of Economics, Chapter 1, Cambridge University Press, Cambridge. Dopfer, K., Potts, J., and Pyka, A. (2017) “Upward and Downward Complementarity: The Meso Core of Evolutionary Growth Theory,” Foundations of Economic Change, Springer, Berlin. Durkheim, E. (1893, 1997) The Division of Labor in Society, Free Press, New York. Elvin, M. (1973) The Pattern of Chinese Past, Stanford University Press, Stanford. Engels, F. (1880) “Socialism: Utopian and Scientific,” in Tucker, R.C. ed. The Marx-Engels Reader, pp. 683–717, 2rd Ed. Norton, New York, 1978. Engels, F. (1884) The Origin of the Family, Private Property, and State, Penguin Classics, 2010. Fama, E.F. (1970) “Efficient Capital Markets: A Review of Theory and Empirical Work,” Journal of Finance, 25(2), 384–433. Ferguson, N. (2002) The Cash Nexus, Money and Power in Modern World, 1700–2000, Basic Books. Field, R.J., and R.M. Noyes (1974) “Oscillations in chemical systems. IV. Limit cycle behavior in a model of a real chemical reaction,” Journal of Chemical Physics, 60 (5), 1877–1884. Friedman, M. (1969) The Optimum Quantity of Money and Other Essays, Aldine, Chicago. Friedman, M. (1992) Money Mischief, HBJ, New York. Galbraith, J.K. (1972) The New Industrial State, 3rd Ed. Penguin. Georgescu-Roegen, N. (1971) The Entropy Law and Economic Process, Harvard University Press, MA: Cambridge. Gowdy, J. (1997) Limited Wants, Unlimited Means: A Reader on Hunter-Gather Economics and the Environment, Island Press. Harris, M. (1978) Cows, Pigs, Wars, and Witches: the Riddles of Culture, Vintage Books. Harris, M. (2006) Cultural Anthropology, 7th Edition, Allyn & Bacon. Hayek, F.A. (1944) The Road to Serfdom, University of Chicago Press. Hayek, F.A. (1945) “The Use of Knowledge in Society,” American Economic Review, 35(4), 519–530. Hayek, F.A. (1975) Monetary Theory and the Trade Cycle, Kelley, New York. Hayek, F.A. (1979) A Tiger by the Tail: The Keynesian Legacy of Inflation, CATO Institute, San Francisco. Hayek, F.A. (1990) Denationalisation of Money-- the Argument Refined: an Analysis of the Theory and Practice of Concurrent Currencies, 3rd ed. Institute of Economic Affairs, London. Heinrichs, H. (2013). “Sharing Economy: A Potential New Pathway to Sustainability,” Gaia, 22 (4): 228–231. Kelton, S. (2020) The Deficit Myth: Modern Monetary Theory and the Birth of the People’s Economy, Public Affairs, New York. Keynes, J.M. (1919) The Economic Consequence of Peace, Macmillan & Co, London. Keynes, J.M. (1936) The General Theory of Employment, Investment, and Money, Macmillian, London. Keynes, J.M. (1940) How to Pay for the War? Harcourt, New York. Kindleberger, C.P. (1986) The World in Depression, 1929-1939, Revised and enlarged edition, University of California Press. Klapp, M.G. (1982) “The State – Landlord or Entrepreneur?” International Organization, 36(3), 575–607. Knight, F.H. (1921) Risk, Uncertainty, and Profit, Houghton Mifflin, New York. Kornai, J. (1980) Economics of Shortage, University of Stockholm, Stockholm.

430

Division of labor as co-evolutionary process Linnaeus, C. (1735) Systema Naturae 1735, Coronet Books Inc, 1964. Lucas, R.E. Jr. (1972) “Expectations and the Neutrality of Money,” Journal of Economic Theory, 4, 103–124. Malthus, T.R. (1798) An Essay on the Principle of Population, London (1798, 2008), Oxford World’s Classics, Oxford University Press, Oxford, 2008. Marx, Karl. (1867) Das Kapital, Verlag von Otto Meissner, Penguin Books, Hamburg, Capital, 1992. Marx, K., and Engels, F. (1848) Communist Manifesto, Penguim, 2002. Maynard Smith, J. (1974) Models in Ecology, Cambridge University Press, London. Mendeleev, D.I. (1869) Mendeleev on the Periodic Law: Selected Writings, 1869–1905, Dover Books, 2013. Muth, J.F. (1961) “Rational Expectations and The Theory of Price Movements,” Econometrica, 29(3), 315–335. Needham, J. (1954) Science and Civilization in China, Vol. I, Cambridge University Press, Cambridge. Nicolis, G., and Prigogine, I. (1977) Self-Organization in Nonequilibrium Systems: From Dissipative Structures to Order through Fluctuations, Wiley, New York. Pauling, L. (1970) Introduction to Quantum Mechanics with Applications to Chemistry, Dover 1988. Piaget, J. (1971) Genetic Epistemology, Norton, New York. Plato (2016) Laws, Cambridge University Press, Cambridge. Popper, K. (1934) The Logic of Scientific Discovery, Hutchinson & Co, London, 1959. Popper, K. (1963) The Open Society and Its Enemy, Princeton University Press, NJ: Princeton. Prigogine, I. (1978) “Time, Structure, and Fluctuations,” Science, 201(4358), 777–785. Prigogine, I. (1980) From Being to Becoming: Time and Complexity in the Physical Sciences, Freeman, San Francisco. Prigogine, I. (1996) The End of Certainty - Time, Chaos, and the New Laws of Nature, Free Press, New York. Prigogine, I., and Lefever, R. (1968) “Symmetry Breaking Instabilities in Dissipative Systems, II,” Journal of Chemical Physics, 48(4), 1695- 1700. Prigogine, I., Nicolis, G., and Babloyantz, A. (1972) “Thermodynamics of Evolution”, PhysicsToday, 25(11), 23–28; 25 (12), 38-44. Rickards, J. (2012) Currency Wars: The Making of the Next Global Crisis, Portfolio. Romer, P.M. (1986) “Increasing Returns and Long-Run Growth,” Journal of Political Economy, 94, 1002–1038. Rostow, W.W. (1990) The Stages of Economic Growth, 3rd Ed. Cambridge University Press, Cambridge. Ryan, J. (2010) A History of the Internet and the Digital Future, Reaktion Books, London. Schrödinger, E. (1948) What is Life? Cambridge University Press, Cambridge. Schumpeter, J.A. (1909) “On the Concept of Social Value,” Quarterly Journal of Economics, 23(2), pp. 213–232. Schumpeter, J.A. (1950) Capitalism, Socialism and Democracy, 3rd ed., Harper, New York. Shleifer, A., and Summers, L.H. (1990) “The Noise Trader Approach to Finance,” Journal of Economic Perspectives, 4(2), 19–33. Smith, A. (1776) The Wealth of Nations, An Inquiry into the Nature and Causes of the Wealth of Nations, University of Chicago Press, 1977. Solow, R.M. (1970) Growth Theory: An Exposition, Oxford University Press, Oxford. Sraffa, P. (1960) Production of Commodities by Means of Commodities, Cambridge University Press, Cambridge. Stigler, G.J. (1951) “The Division of Labor Is Limited by the Extent of the Market,” Journal of Political Economy, 59, 185–193. Stillman, R. (1986). Dow Jones Industrial Average: History and Role in an Investment Strategy, Dow Jones-Irwin, Homewood, Ill. Tang, Y., and Chen, P. (2014) “Time Varying Moments, Regime Switch, and Crisis Warning: The Birth-Death Process with Changing Transition Probability,” Physica A, 404, 56–64. Tang, Y., and Chen, P. (2015) “Transition Probability, Dynamic Regimes, and the Critical Point of Financial Crisis,” Physica A, 430, 11–20. Thaler, R.H., and Sunstein, C.R. (2021) Nudge, Final Ed.Yale University Press. Toffler, A. (1980) The Third Wave, William Morrow, New York. UNSTATS (2022) https://unstats.un.org/unsd/snaama/Basic, data retrieved on Mar. 6.

431

Ping Chen Veblen, T. (1898) “Why Is Economics Not an Evolutionary Science,” Quarterly Journal of Economics, 12(4), 373–397. Veblen, T. (1899) The Theory of the Leisure Class: An Economic Study of Institutions. Penguin, New York, 1994. Wallis, J.J., and North, D.C. (1986) “Measuring the Transaction Sector in the American Economy, 1870–1970,” pp. 95–148, and “Comments” by Lance E. Davis, pp.149-161, in Engerman, S.L. and Gallman, R.E. eds., Long-term factors in American economic growth, Studies on income and wealth, vol. 51 (National Bureau of Economic Research), University of Chicago Press, Chicago. Weber, M. (1905) Economy and Society: An Outline of Interpretive Sociology, Univ. of Cal. Press, 1978. Wilson, E.O. (1975) Sociobiology: The New Synthesis, Harvard University Press.

432

34 EVOLUTIONARY ECONOMICS AND LDCS An African perspective J. Fagerberg, E. Kraemer-Mbula, and E. Lorenz

34.1

Introduction

What mainly distinguishes evolutionary economics from other approaches to the study of economic phenomena is its emphasis on qualitative changes in historical time, and factors influencing such processes. Since qualitative changes in production, consumption, organizational forms, institutions, etc. arguably are what economic development is about, one would expect an approach focussing on such changes to be very relevant for the study of what policymakers in low-income countries can do to upgrade economic structures and increase the welfare of the population. Nevertheless, evolutionary economists from Joseph Schumpeter onwards have mainly focused on the leading capitalist countries and other highly mature economies. The small set of (mostly Asian) countries that during the last half-century managed to substantially reduce the gap in productivity and income vis a vis the developed part of the world has also received attention. However, very little systematic work has been undertaken on the economics of lower-income countries from an evolutionary perspective. This paper aims to address this gap, by explicitly focusing on the extent to which insights from evolutionary economics may enrich our understanding of the prospects for lower-income countries in Africa. As a prelude, Section 34.2 below discusses some central insights from the evolutionary economics literature that potentially may be of high relevance for the task. The outcome of this exercise is put to the test in Sections 34.3 and 34.4 based on evidence from two low-income African countries, Kenya and Rwanda, with Section 34.3 focusing on the impact of recent technological changes, and Section 34.4 on the role of policy in promoting these changes. Section 34.5 sums up the lessons.

34.2

Economic development in low-income countries: An evolutionary perspective

If the hallmark of neoclassical economics is equilibrium (and how to achieve and sustain it), evolutionary economics is about qualitative, innovation-driven change. It was Joseph Schumpeter who more than a century ago started to analyze economic development in this way (Fagerberg 2003). From this perspective, the key social phenomenon that needs to be

DOI: 10.4324/9780429398971-37

433

J. Fagerberg et al.

understood is innovation, defined as the introduction of novelty in the economic sphere (rather than, say, as new ideas or “invention”).1 While it is common to associate innovation with outstanding scientific breakthroughs and/or high-tech environments, Schumpeter – and evolutionary economics – places innovation in a much broader context. According to Schumpeter, innovations come in many different shapes, e.g., not only technological but also organizational, and different sizes, ranging from very radical innovations to minor changes in existing products and processes, and they all matter. Moreover, innovation is not something that only goes on in select hightech environments or the manufacturing industry. Innovation, from an evolutionary perspective, is something that goes on – and matters – in all kinds of economic activities, i.e., in services and industry, as well as in the public and private sectors (Fagerberg 2004). Although much innovation occurs in private businesses, and with a profit motive, innovations may also occur in other settings and be driven by other motivations, e.g., so-called social innovations (Moulaert et al. 2013). Furthermore, novelty may be regarded as contextdependent, i.e., introducing something for the first time in a new context may also qualify as an innovation, even if it is not necessarily “new to the world” (Smith 2004). Thus, evolutionary economics sees innovation as a potent force for change in a broad range of sectors and activities, in developed as well as developing countries. This being said, Schumpeter and many evolutionary economists with him have an especially keen interest in radical innovations, particularly those having a major influence on the behavior of the entire global economy over an extended period of time, what Freeman and Perez (1988) call “technological revolutions” or “changes in techno-economic paradigms”. The defining feature, they argue, is the existence of a cheap key input characterized by rapidly declining costs, almost unlimited supply, and very broad applicability (ibid, p. 48). This may lead to a virtuous circle, in which both the industry producing the key input and industries using it extensively (the “carrier” branches) grow very fast, resulting in rapid productivity growth and extensive structural changes in the economy. Examples of such key inputs are oil (from around 1900 onwards) and microelectronics during the last 50 years. More recently it has been argued that the rapid progress in renewable energy technologies, particularly solar and wind, also qualify as a technological revolution and can be expected to have a similarly broad impact all over the globe (Mathews 2013, 2014). A parallel point has been made regarding the potentially transformative effects of the so-called digital revolution of the last 10 to 15 years, driven by mobile telephony, rising internet usage and the uptake of digital service platforms (Zysman and Kenney 2016; Africa’s Pulse 2019). As Freeman and Perez (1988) emphasize, technological revolutions have implications far beyond technical change, involving large infrastructural investments with pervasive effects throughout the economy, changes in managerial and organization practices, a new skill profile for the workforce, and changes in patterns of distribution and consumption of goods and services. Technological revolutions imply challenges as well as opportunities. But in current times for developing countries, the opportunities may be what strikes the eye. For example, in the developing part of the world, the transition from wired to mobile telephony made investments in costly fixed landline telecommunications infrastructure unnecessary and paved the way for making services available through digital platforms to people who would otherwise have been excluded from them. Much in the same way the ongoing progress in renewable solar and wind energy technology increasingly allows some rural populations without access to the national grid to benefit from small scale and decentralized renewable energy installations. Digitalization may support the adoption of renewable energy in several ways; for 434

Evolutionary economics and LDCs

example, digital technologies can transform value chain relations supporting finance, installation and maintenance of standalone off-grid solar providing electricity on a decentralized basis.2 Digital data analytics may also be applied to mini-grids powered by renewable resources to balance electricity demand and supply and ensure efficient system operation (Fritzsche et al. 2019). Hence, the interaction between the renewable energy revolution and the digital revolution may offer great opportunities for developing countries, see the discussion in Sections 34.3–34.4 of this chapter. Innovation is increasingly acknowledged – not only by evolutionary economists but more broadly – as a key factor in economic development (Fagerberg et al. 2010, Kraemer-Mbula and Wamae 2010) – affecting technology-intensive activities (Lorenz and Kraemer-Mbula 2020), as well as less sophisticated ones (Kraemer-Mbula et al. 2019). Innovation is also seen as a tool for dealing with more specific challenges that policymakers are facing (giving rise to socalled mission-oriented innovation policies, see, e.g., Mazzucato 2013) as well as broader developmental challenges related to environmental sustainability and social inclusion (such as transformative innovation policies, see, e.g. Steward 2012, Schot and Steinmueller 2018). It is not surprising, therefore, that attention to the role of governance and policy in encouraging and influencing innovation has been on the rise during the last few decades (Fagerberg 2017). However, the great uncertainty in innovation, as well as the widely distributed nature of relevant knowledge, has generally led evolutionary economists to emphasize policies that strengthen the national system’s ability to innovate, such as supporting capability-building, financial access and interaction between different actors (e.g., public and private), rather than devising more specific paths for how the future should look like. Nevertheless, the global climate challenge (Stern 2015) and the political reactions to it (e.g., the UN’s Paris convention from 2015), have arguably provided a clearer direction for society’s development in the years to come, e.g., a transition from an energy system based on burning fossil fuels to a system based on renewable energy and huge savings in energy and resource use (circular economy). Hence, an important question that has attracted much attention recently is how innovation – and the policies supporting it – can contribute to these changes.3 Among the requirements mentioned in the literature are adequate policy direction, improved coordination between different parts of government (whose activities matter for innovation) and greater involvement of multiple stakeholders in society. While undoubtedly very challenging for the already developed part of the world, deeply embedded as they are in the old fossilfuel-based industrial system, developing countries may possibly turn this situation to their advantage, by – as pointed out above – embracing the opportunities offered by the more recent technological developments in renewable energy in interaction with those in ICTs involving the use of digital communications and media infrastructures.

34.3

Windows of opportunity for transformative change in Kenya and Rwanda

Achieving Africa’s ambitious vision towards a sustainable and inclusive future (as set by the African Union’s pan-African strategy, Agenda 2063: The Africa We Want), requires profound transformations of production, consumption and governance systems. Transforming Africa into “a global powerhouse of the future”, as this strategy envisions, requires not only catching up through imitation but widespread innovation, with new products, new services, new processes and new businesses emerging, that result in sustainable outcomes. Already, offgrid renewables are enabling “unscaling” of the energy system, challenging the monopoly power of national utilities through the emergence in Africa of ‘prosumers’, i.e., people who are 435

J. Fagerberg et al.

both producers and consumers of power. The use of mobile money has grown exponentially in paving the way to develop indigenous digital service platforms transforming patterns of production and consumption (Kaplinsky and Kraemer-Mbula 2022). These transformations may not be merely avenues to “catch-up”, following the path of already mature economies. One may also see them as opportunities for “path-breaking”, as an essential strategy in the context of climate change and the urgent need to reverse the unequal patterns of distribution from earlier development paths. But transformation is a long-term process; not automatic, and depends on the strength of the national innovation system, shaped by the interactions and learning among various organizations and institutions under the influence of government policies. It also depends on having the foresight and political will to put in place these policies. In this regard, the commitment to accelerate the continent’s transformation by harnessing its vast renewable energy potential and digital technologies is reflected in current strategies such as the Africa Renewable Energy Initiative launched in 2015, and the Digital Transformation Strategy for Africa (2020). Below we focus on the cases of Kenya and Rwanda, two low-income African countries that stand out not only for being among the top ten fastest-growing countries in Africa since 2000 (World Bank Development Indicators)4 but also for pursuing from the early 2000s policies aiming to drive transformative change (see section 34.4 below).

34.3.1

Digital and energy transformation in Kenya and Rwanda

A key component of the digital transformation underway in Kenya and Rwanda is the rapid development of wireless mobile cellular communication networks as an alternative to making costly investments in wired landline communication systems. Table 34.1 shows that, as with most countries in Sub-Saharan Africa (SSA), neither Kenya nor Rwanda made significant investments in the fixed landline telecommunication systems that were infrastructural supports for the ICT revolution taking hold in higher-income countries from the 1980s. The feeble development of fixed telephone lines paralleled the lack of internet use in Kenya and Rwanda, since at that time an internet connection required a computer and the use of a modem to dial up a connection. Table 34.1 Number of Fixed Line Telephone Subscriptions per 100 Persons Year

Kenya

Rwanda

SSA

High-Income Countries

1990 2000 2010 2018

0.74 0.92 0.91 0.13

0.14 0.22 0.39 0.11

0.99 1.38 1.48 0.81

40.53 53.90 46.93 38.80

Source: World Bank Data: https://data.worldbank.org/indicator/IT.MLT.MAIN.P2

The number of fixed-line subscriptions in high-income countries declined after 2000 and was quickly surpassed by mobile subscriptions. The radical transformation in telecommunications access in Kenya and Rwanda brought about by the diffusion of mobile phones can be seen in Figure 34.1 below. Mobile telephone subscriptions increased rapidly after 2005 substantially closing the gap with high-income countries by 2019. This diffusion improved internet access supported by the dramatic decline in broadband cost after 2009 with the arrival of several submarine cables connecting the African continent to global internet services. In 2019 over 48% of Kenya’s population had access to broadband 436

Evolutionary economics and LDCs 150 High-income Kenya

100

SSA Rwanda 50

0 2000

Figure 34.1

2005

2010 Time

2015

2020

Number of cellular mobile subscriptions per 100 persons.

Source: https://data.worldbank.org/indicator/IT.CEL.SETS.P2.

connectivity, predominately through the mobile phone network (World Bank Group 2019). In Rwanda, access to international bandwidth increased tenfold between 2015 and 2020 giving Rwanda the highest 4G coverage of any eastern African country, standing officially at over 95% of the population (World Bank Group 2020). These technological and infrastructural developments, as pointed out above, can only be expected to have a transformative effect on the economy if they are accompanied by innovations in business models and patterns of production and consumption. In Kenya and Rwanda, this manifested in the development of indigenous digital platforms providing access to a range of new services including finance, health, and agricultural supply chain management. Kenya is thought to have experienced the fastest growth in the number of digital platforms in SSA (estimated 118 platforms in 2019), mostly indigenous or homegrown (80% of them) (Insight2impact 2020). The most widely discussed case is undoubtedly Safaricom’s mobile money platform, MPesa, which experienced explosive growth from 2 million registered users within a year of its start in 2007 to over 10 million by the end of 2010 (Demirguc-Kunt et al. 2015), and over 40 million worldwide in 2020. The use of mobile money expanded from making remittances to include individual payments for goods and services and business transactions, including payments for inputs, paying employees, and receiving customer payments (Gosavi 2015; Lorenz and Pommet 2021). Further, the platform spawns a range of innovative digital services based on the Pay-As-You-Go business model allowing customers to use their mobile money accounts to finance their purchase of assets over time, including solar home systems (SHS) and solar-powered televisions (Adwek et al. 2020). The uptake of mobile money in Rwanda, while significant, has been less rapid than in Kenya; with 31% of Rwandan adults having a mobile money account in 2017, up from 18% in 2014. As in the case of Kenya, remittances have been a significant use of mobile money with approximately 33% of the adult population sending or receiving remittances in 2017, and of these about 73% using mobile money to make the transaction.5 In Rwanda, the development of digital platforms has to a greater extent than in Kenya been driven by public investments. The Rwandan government has pursued an ambitious e-government strategy expanding its e-services from only five in 2015 to some 89 in 2018 (World Bank Group 2020, p. 25). 437

J. Fagerberg et al.

The transformative impact of new digital technologies is paralleled by renewable energy’s potential to transform economic activity, especially in rural areas, by providing cheap and decentralized energy supply. The share of the rural population with access to electricity increased in Kenya from about 30% in 2011 to 72% in 2018 thanks to a combination of grid extensions and investments in off-grid capacity. In Rwanda, progress was slower (23% of the rural population had access in 2018).6 While solar energy constitutes a small share of power generation capacity in both countries, solar in the form of off-grid mini and solar home systems plays an important role in rural electrification and accounts for most of the recent increase in off-grid capacity (see Figure 34.2 below). The uptake of off-grid solar has been driven by the steep decline in the cost of solar photovoltaics, estimated at 82% between 2010 and 2019, more than any other electricity generation technology (IRENA 2020b). The adoption of solar home systems (SHS) and solar lamps has largely depended on digital platform-based start-ups that offer individual households solar energy systems that are paid for with mobile money using pay-as-you-go financing. In Kenya, M-Kopa is the most important provider of off-grid solar home systems having wired by 2018 over 600,000 homes in East Africa.7 Most of the rural population serviced by M-Kopa depends on agriculture, and there is an emerging market for solar-powered appliances in agriculture and downstream agro-processing, including solar pumps and solarpowered milling and pressing equipment. As in the case of solar home systems, this equipment may be financed with pay-as-you-go systems using the M-Pesa mobile money platform (Africa’s Pulse 2019) (Figure 34.2). In Rwanda, linkages between digital platforms and off-grid solar are emerging. In 2018, approximately 11% of the population was connected to off-grid systems, primarily solarbased. Several independent companies have wired up to 300,000 households with solar home systems and in some cases offer pay-as-you-go finance through mobile phone platforms (Rwanda Ministry of Infrastructure, 2017b).8 As in Kenya, there is an emerging market for solar-powered agricultural equipment (possibly financed through pay-as-you-go financing systems).9 Kenya: total off-grid

50

Megawatts

40

Kenya: solar off-grid

30

Rwanda: total off-grid

20 Rwanda; solar off-grid

10 0 2010

Figure 34.2

2012

2014 Year

2016

2018

Off-grid energy capacity in Kenya and Rwanda in Megawatts.

Source: IRENA (2020a).

438

Evolutionary economics and LDCs

These developments in standalone off-grid solar home systems and other solar-powered products hint at the transformative potential of solar energy. They indicate that solar is more than an infrastructure for energy generation and supply. Solar energy, facilitated by digital mobile finance platforms, demonstrates its transformative potential by being incorporated in many products in several sectors.

34.4

Policies for transformative change

A profound transformation is required for Africa to achieve its vision while responding to the urgent threats posed by climate change and the social consequences of the unequal distribution of wealth and opportunities. Arguably, a poorly managed transformation is unlikely to result in the realization of the “path-breaking” potential brought by renewable energies and new digital technologies. Transformative change hinges on the actions of many stakeholders, including not only government and private sector, but also NGOs, international partners and civil society, among others. Governments can do much to support these transformations through appropriate policies, regulations, incentive mechanisms, and supporting infrastructure; and by ensuring that innovative firms can mobilize the necessary resources (i.e. skills, finance, technologies, etc.) to roll out their innovations and connect them to potential markets. Below we discuss, for Kenya and Rwanda, the government’s role in promoting transformative change through regulation, infrastructural investment, skills development and support to entrepreneurship.

34.4.1

Governance and regulatory environment for the digital economy

In Kenya and Rwanda, government policies have been guided since the early 2000s by longterm visions for achieving higher growth with greater inclusiveness. Rwanda’s Vision 2020 established in 2000 aimed to transform Rwanda from an agrarian to a knowledge-based society with universal access to education and healthcare and promoting private-sector-led development as pillars. Successive 5-year development plans attributed a leading and crosscutting role to ICTs (Government of Rwanda 2001). Kenya’s Vision 2030 envisages a globally competitive and prosperous economy and society, simultaneously pursuing economic growth and social inclusiveness. From 2008, a series of medium-term plans identify as key foundations the development of energy and ICT infrastructure, and science, technology and innovation (STI) capabilities (Republic of Kenya 2008). To encourage private-sector development, the Kenyan and Rwandan governments moved early on to liberalize their telecommunications sectors. However, in practice, the monopoly over mobile telephone and internet exercised by the state-owned providers only came to an end in 2006 in Rwanda and in 2007 in Kenya (Mureithi, 2017; Government of Rwanda 2005). In Kenya, a key development was the Central Bank’s decision in 2007 to issue Safaricom a letter of no objection authorizing it to launch M-Pesa, enabling the regulatory environment for the subsequent explosive growth of mobile money.

34.4.2 Infrastructural investments for digital communications and renewable energies Internet connectivity improved substantially from 2009 in both countries with the landing of several international submarine cables as described above. This was complemented by government investments in domestic fiber-optic infrastructures to connect different regions 439

J. Fagerberg et al.

and cities. In Kenya, there have been substantial private sector infrastructural investments by the main mobile network operators (Orange Telkom, Safaricom, Airtel and Essar) who have developed their own ICT infrastructures (Kenya Master ICT Plan 2014, p. 31). In Rwanda, public investments made in the national and metropolitan fiber networks are open to private enterprises in an attempt to promote private sector investment (Republic of Rwanda 2017a). A primary objective of energy policy in both countries has been to reduce dependence on biomass, which presents severe health and environmental risks. Policies in Kenya to increase access to electricity through renewables focused mainly on national grid extensions until 2016–17, when greater emphasis was given to solar off-grid with an NGO supported program for the installation of solar PV systems in primary and secondary schools as well as health and administrative centers.10 Off-grid solar was given a boost by the creation of the Rural Electrification and Renewable Energy Corporation (REREC) in 2019 with a mandate to put renewable energy, including mini-grids, standalone solar systems, and solar water pumps for community facilities, at the center of policy.11 In Rwanda, off-grid renewables are central in the plan to achieve universal electricity access by 2024. The 2018 Energy Sector Strategic Plan envisages rough parity in the shares of the population serviced by grid and off-grid energy power generation, with solar home systems set to play a major role (Republic of Rwanda 2017b). Financial support for lowincome households and communities to access off-grid solar energy is provided through the provisions of the 2016 Rural Electrification Strategy.

34.4.3 Policies for skills development Reflecting the key role of the ICT sector in Rwanda’s development strategy, government policy from the early 2000s acted to diffuse ICTs within the educational system from primary school upwards. By 2017, 44% of primary and 60% of secondary schools had access to ICT for teaching and learning (World Bank Group 2019). At the higher education level, several ICT training programs were established in major universities and technical institutes, and between 2002 and 2005 around 2000 professionals in engineering, technology and computer science/ICTs received degrees, mainly at the BSc and diploma levels. (Government of Rwanda 2005, p. 32). Education policies in Kenya have also emphasized digital skills development. The 2005 Kenya Education Sector Support Program featured ICT as one of the priority areas, aiming to mainstream ICTs into curriculum and teaching. A 2020 World Bank group report notes that policies and programs promoting the use of ICT for teaching and learning, including a competency-based framework that features digital skills, are formally in place. A key flagship initiative at the primary level was the 2016 Digital Literacy Program providing connectivity, devices, and electricity. However, while over 90% of primary schools are covered by the initiative, the report noted that only 36% of schools use the equipment as intended (World Bank Group 2019).

34.4.4

Multi-stakeholder support for the entrepreneurship ecosystem

Kenya is known for its vibrant digital innovation ecosystem with multiple innovation hubs spawning several hundred digitally anchored start-ups (World Bank Group 2019). However, government support for start-ups in the form of subsidies or tax incentives has not been a major driver. A constellation of various actors has fostered Kenya’s fast-growing start-up 440

Evolutionary economics and LDCs

landscape, including venture capital, global tech giants (such as Intel, Google and Facebook), international donors, supporting incubators, hubs and accelerators. Successive national plans have emphasized entrepreneurship training linked to the objectives of increasing youth employment and women’s empowerment.12 Nevertheless, the Start-up Bill proposed in 2020 breaks with this pattern by including tax incentives for those start-ups that are majority Kenyan-owned and focused on innovation.13 Despite the emphasis given to private sector-led development in Rwanda, the government has taken the leading role in driving the digital economy through its ambitious e-government strategy (World Bank Group 2020). Private sector investment in the energy infrastructure has been promoted largely at government initiative through public-private partnerships to provide rural access to off-grid solar energy (Republic of Rwanda 2017b). Some limited success in promoting dynamic start-ups and entrepreneurship has been achieved through the support of NGOs and international organizations, and a number of investment funds targeting start-ups have been created with the backing of the World Bank Group or the African Development Bank. Hence, international donors and investors are playing an important role in shaping the Rwandan start-up ecosystem.

34.5 Concluding remarks The economic prospects of low-income countries have conventionally been framed as mechanisms that may support their transition to high-income status through copying technologies and practices in use elsewhere. However, the complex challenges and opportunities facing humankind today, as well as the urgency to move towards more inclusive and sustainable modes of development, point to the limitations of adopting such a lens. This chapter argues that evolutionary economics lends us a more suitable set of tools to explore “path-breaking” modalities of development, relying in particular on the transformative power of digital technologies and renewable energies, the importance of a national vision and the institutional framework steering the direction of change as well as actions of multiple actors and stakeholders comprising the innovation and entrepreneurship ecosystems, which expand beyond national borders. In this chapter, we particularly highlight the central role that governments can play in triggering such changes through the mobilization a diverse set of actors in the path towards transformative change. As the examples of Kenya and Rwanda indicate, some low-income countries have already recognized the opportunities ahead, providing a possible point of departure for other developing countries.

BIO Jan Fagerberg Professor Jan Fagerberg is affiliated with the INTRANSIT Center at the University of Oslo. In his research, Fagerberg has among other things focused on the relationship between innovation and development. He has also worked on innovation theory, innovation systems and innovation policy.

BIO Erika Kraemer-Mbula Professor of economics, and chairholder of the DSI/NRF/Newton Fund Trilateral Research Chair in Transformative Innovation, the Fourth Industrial Revolution and Sustainable Development, at the College of Business and Economics, University of Johannesburg, South 441

J. Fagerberg et al.

Africa. She specialises in science, technology and innovation policy analysis and innovation systems in connection to equitable and sustainable development. The work of Prof. Kraemer-Mbula and Prof. Lorenz in this study has been supported by the National Research Foundation of South Africa (Grant Numbers: 118873).

BIO Edward Lorenz Edward Lorenz is Emeritus Professor at the University of Côte d’Azur, France. He also holds the positions of distinguished visiting professor at the University of Johannesburg, South Africa and adjunct professor at Aalborg University, Denmark. His research focuses on the impact of new and emerging technologies on innovation and sustainable development.

Notes 1 As Schumpeter famously pointed out: “As long as they are not carried out into practice, inventions are economically irrelevant. And to carry any improvement into effect is a task entirely different from the inventing of it, and a task, moreover, requiring entirely different kinds of aptitudes” ( Schumpeter 1934, p. 88). 2 See the discussion below of M-KOPA Solar in Kenya and the equivalent pay-as-you-go solar home systems providers in Rwanda. 3 See e.g., Schot and Steinmueller 2018, Fagerberg 2018. 4 All World Bank development indicators cited in this chapter are available at: data.worldbank.org/. 5 See: Global Findex Database, 2017: https://globalfindex.worldbank.org/. 6 See: Republic of Kenya 2018; Republic of Rwanda 2018; World Bank Development Indicators). 7 See: http://www.m-kopa.com/. 8 For the activities of the enterprise Ignite Power in Rwanda, see https://www.esi-africa.com/news/ off-grid-power-illuminates-rwandan-villages/. 9 See, for example, https://futurepump.com/futurepump-rwanda/. 10 See: https://energy4impact.org/news/improving-health-and-education-services-marginalised-ruralcommunities-kenya. 11 See: See: https://www.rerec.co.ke/ for the off-grid programs of the Rural Electrification and Renewable Energy Corporation. 12 Examples include the Youth Enterprise Development Fund (YEDF) to support the training of young entrepreneurs and the UWEZO Fund to support youth employment and women’s empowerment (Republic of Kenya, 2013). 13 For the bill, see https://www.bowmanslaw.com/insights/intellectual-property/kenyas-senateintroduces-the-startup-bill-2020/.

References Adwek, G., Boxiong, S., Ndolo, P.O., …, Yabo, A.C. (2020). The solar energy access in Kenya: A review focusing on Pay-As-You-Go solar home system. Environment, Development and Sustainability 22(5) No 4, pp. 3897–3938. Africa’s Pulse. (2019). ‘An Analysis of Issues Shaping Africa’s Future’, Volume 19, World Bank Group. Demirguc-Kunt, A., Klapper, L., Singer, D., and Van Oudheusden, P. (2015). The Global Findex Database 2014: Measuring Financial Inclusion around the World. Policy Research working paper; no. WPS 7255, World Bank Group. Fagerberg, J. (2003) Schumpeter and the revival of evolutionary economics: An appraisal of the literature. Journal of Evolutionary Economics 13: 125–159. Fagerberg, J. (2004) Innovation: A guide to the Literature. in Fagerberg, J., Mowery, D., and Nelson, R. (eds.) The Oxford Handbook of Innovation, Oxford: Oxford University Press, 2004, pp. 1–26. Fagerberg, J. (2017). Innovation policy: Rationales, lessons and challenges. Journal of Economic Surveys, 31: 497–512.

442

Evolutionary economics and LDCs Fagerberg, J. (2018). Mobilizing innovation for sustainability transitions: A comment on transformative innovation policy. Research Policy, 47(9): 1568–1576. Fagerberg, J., Srholec, M., and Verspagen, B. (2010). Innovation and Economic Development. in Hall, B., and Rosenberg, N. (eds.), Handbook of the Economics of Innovation. Vol. II. North Holland, pp. 833–872. Freeman, C. and Perez, C. (1988). Structural Crises of Adjustment: Business Cycles and Investment Behaviour. in Dosi, G., Freeman, C., Nelson, R.R., Silverberg, G. and Soete, L.G. (eds.), Technical Change and Economic Theory, London: Pinter, pp. 38–66. Fritzsche, K., Shuttleworth, L., Brand, B., and Blechinger, P. (2019). Exploring the nexus of minigrids and digital technologies. Potentials, challenges and options for sustainable energy access in SubSaharan Africa, Potsdam: Institute for Advanced Sustainability Studies (IASS). Gosavi, A. (2015). The next frontier of mobile money usage-firms-evidence from Sub-Saharan Africa. In 19th Annual Western Hemispheric Trade Conference April 15-17, 2015, Laredo, TX, USA, Conference Proceedings (p. 139). Government of Rwanda (2001). An Integrated ICT-led Socio-Economic Development Policy and Plan for Rwanda 2001–2005, Kigali. Government of Rwanda (2005). An Integrated ICT-led Socio-Economic Development Policy and Plan for Rwanda 2006–2010, Kigali. Insight2Impact. (2020). ‘Africa’s Digital Platforms: Overview of emerging market trends’. https:// i2ifacility.org/search?utf8=%E2%9C%93&term=platforms IRENA. (2020b). Renewable Power Generation Costs in 2019. Abu Dhabi: The International Renewable Energy Agency. Kaplinsky, R., & Kraemer-Mbula, E. (2022). Innovation and uneven development: The challenge for low-and middle-income economies. Research Policy, 51(2), 104394. Kenya Master ICT Plan (2014). Towards a Digital Kenya. Nairobi: ICT Authority. Kraemer-Mbula, E. and Wamae, W. (eds.) (2010). Innovation and the Development Agenda. OECD Publishing. Kraemer-Mbula, E., Lorenz, E., Takala-Greenish, L., Jegede, O.O., Garba, T., Mutambala, M., and Esemu, T. (2019). Are African micro-and small enterprises misunderstood? Unpacking the relationship between work organisation, capability development and innovation. International Journal of Technological Learning, Innovation and Development, 11(1): 1–30. Lorenz, E. and Kraemer-Mbula, E. (2020). The Impacts of Adopting 4ir-related Technologies on Employment and Skills: The Case of the Automotive and Mining Equipment Manufacturers In South Africa, in Mazibuko-Makena, Z. and Kraemer-Mbula, E. (eds) Leap 4.0: African Perspectives on the Fourth Industrial Revolution, Johannesburg: Jacana Publishers. Lorenz, E., & Pommet, S. (2021). Mobile money, inclusive finance and enterprise innovativeness: an analysis of East African nations. Industry and Innovation, Taylor & Francis Journals, 28(2): 136–159, February. Mathews, J.A. (2013). The renewable energies technology surge: A new techno-economic paradigm in the making?, in Futures, 46: 10–22. Mathews, J.A. (2014). Greening of Capitalism: How Asia Is Driving the Next Great Transformation. Stanford: Stanford University Press. Mazzucato, M. (2013). The entrepreneurial state: debunking private vs. public sector myths. London, UK: Anthem Press. Moulaert, F., MacCallum, D., Mehmood, A., and Hamdouch, A. (2013). The International Handbook on Social Innovation: Collective Action, Social Learning and Transdisciplinary Research. Cheltenham: Edward Elgar. Mureithi, M. (2017). The Internet Journey for Kenya: The Interplay of Disruptive Innovation and Entrepreneurship in Fueling Rapid Growth. In Ndemo, B., and Weiss, T (eds), Digital Kenya: An Entrepreneurial Revolution in the Making, Palgrave: MacMillan, 27–53. Republic of Kenya. (2008). First Medium-Term Plan, 2008-2013. Office of the Prime Minister. Republic of Rwanda. (2017a). Ministry of Information Technology and Communications. ICT Sector Strategic Plan, 2018–2024. Republic of Rwanda (2017b). Ministry of Infrastructure, Draft Energy Strategic Sector Plan 2018/19 to 2023–24.

443

J. Fagerberg et al. Schot, J. and Steinmueller, W.E. (2018) Three frames for innovation policy: R&D, systems of innovation and transformative change. Research Policy, 47(9): 1554–1567. Schumpeter, J.A. (1934). The Theory of Economic Development. Cambridge, Mass: Harvard University Press. Smith, K. (2004). Measuring Innovation. in Fagerberg, J., Mowery, D., and Nelson, R. (eds.), The Oxford Handbook of Innovation, Oxford: Oxford University Press, pp. 148–178. Stern, N. (2015). Why are We Waiting? The Logic, Urgency, and Promise of Tackling Climate Change. Cambridge: The MIT Press. Steward, F. (2012). Transformative innovation policy to meet the challenge of climate change: sociotechnical networks aligned with consumption and end-use as new transition arenas for a lowcarbon society or green economy. Technology Analysis & Strategic Management, 24(4), 331–343. World Bank Group. (2019). Securing Future Growth, Policies to Kenya’s Digital Transformation. Economic Update, No. 20. World Bank Group. (2020). Accelerating digital transformation. Rwanda Economic Update, No. 20. Zysman, J. and Kenney, M. (2016). The next phase in the digital revolution: Platforms, abundant computing, growth and employment. ETLA Reports, The Research Institute of the Finnish Economy, 61, 1–26.

444

35 GLOBALIZATION AND ITS GOVERNANCE IN AN EVOLUTIONARY PERSPECTIVE Pascal Petit

From an economic point of view (according to the UN Committee for Development Policy), globalization can be defined as a process of increasing interdependence of world economies resulting from the growing scale of cross-border trade of commodities and services, flow of international capital and wide and rapid spread of technologies, all of which implying a growing common significance of information in all types of productive activities and marketization. The process is thus multidimensional and may concern differently countries along time and space. Its governance is thus bound to vary with the international accords and power relations prevailing among the set of countries under view. A priori there is no reason for the process of globalization to be monotonous and all the more so that it is multidimensional. At some periods though, there is such rapid broad expansion of this process that observers speak of waves of globalization. Richard Baldwin, expert in international economics, in his 2016 book On the great convergence distinguishes thus two waves of globalization. The first was linked to the reduction in transportation costs brought by the industrial revolution (with the successive diffusion of steam power, electrification and fuel engines). The second wave of globalization begun at the end of the 20th century with the reduction in communication costs brought by the diffusion of ICTs (information and communication technologies). The first wave led to a continuous development of international trade in products made in different places of the globe. The second phase paved the way to an international division of production chains (often referred to as global value chains). Baldwin (2016) even suggested that the development of ICTs (information and communication technologies), whereby robots could be monitored from anywhere in the planet, could represent a third wave of globalization in reducing the cost of face to face communications. Though this evolutionary perspective where technical changes are playing a key role leaves open diverse alternatives. Clearly these three “unbundlings” (to take Baldwin’s word) of the conditions that tied, in time and space, production and consumption refer to contemporary periods of times where experiences may be very different and changing for the countries under view. Issues of power relations, of food or sanitary security have to be accounted for, as they are at the root of the international processes under view. Such clarification is all the more necessary that many social scientists speak of globalization, DOI: 10.4324/9780429398971-38

445

Pascal Petit

even in earlier periods, to evoke the military expansion of empires or the maritime expeditions to discover the world and develop some specific trades (in spices or precious minerals). Baldwin (2006) himself thus speaks of four phases of globalization to distinguish two earlier periods when a) modern humans left Africa and colonized the world and b) when climate warming and stabilizing allowed development of agriculture and rise of ancient civilizations (with military conquests and religious proselytisms). These two early phases of international exchanges are clearly very distinct from the two last phases referred above by Baldwin as “waves” of globalization. Interestingly part of these past changes in globalization already involved environmental changes. The long retrospective view of globalization given by Baldwin is interesting but incomplete if one tries to understand the governance of these globalization movements. One needs in that respect to know more on some governing principles which both motivated and legitimized the development of international relations. One has to do with the “modern” status of nation states as it emerged in the seventeen century, a status which is still so present that one tends to forget to question its path dependency evolution. The other, in an even longer perspective, has to do with the ideological representation one has of mankind in their planet in a vast surrounding space. In all the four phases of globalization previously evoked, religion played its part, as well as more generally the conceptions people had of the universe and of our planet. Galileo played his part in the 17th century and all along, till our days, the discourse of sciences help to frame our view of the world and thus of the context of international relations. It is all the more important to keep that in mind that the big potential that Baldwin sees for a third wave of globalization, following the development of ICTs and AI (artificial intelligence) strongly depends on our ability to monitor these scientific developments. On this line, the scientific contemporary acknowledgment of the major importance of our relation with nature comes as a determining new Galilean revolution as we shall see later in this chapter. Let us first come back on the major stage in our apprehension of internationalization that the making of nation state constitutes. To analyze processes taking place at a global scale, involving potentially different number of countries and developing international ties in different ways, one needs to start from a systemic view of the world wide set of countries. The break in that respect is given by the treaty of Westphalia (1648) which can be seen as the benchmark of the emergence of the modern system of sovereign states. To settle the peace between France and Germany, following a religious war of some 30 years, the treaty established that independent states should not interfere in each other’s domestic affairs. This Westphalian principle of noninterference in other countries affairs became commonly accepted by the mid-18th century (see Krasner 2009) and was enhanced afterwards by the rise in the 19th century of nationalist currents, trying to associate as much as possible states with nations- groups of people, united by language and culture. This Westphalian order, affirming the sovereignty of states, constitutes in effect a benchmark to appreciate the impacts of the many international transactions in which states are involved. Within this framework, the question is then to see how the diversity of international relations is dealt with. In other words, the issue is to know to what extent the governance of these international relations is compatible with the Westphalian principle of state sovereignty. Most of the international relations under view take place in legal contexts, defined among varying sets of countries. But countries have not at all the same power to enter in such international legal arrangements and some have to realistically adjust accordingly between gains and losses. Much depends on the geography, considering access facilities and on history, following past experiences of 446

Globalization and its governance in an evolutionary perspective

relations. Levels of development also matter to determine the positions of countries in these networks of international accords. The very idea of sovereignty means that each state unites around broad objectives to meet challenges which may change along time, considering its experiences in terms of external relations. A major issue in that respect is its military security. This issue is clearly path dependent, being marked by a long list of conflicts and alliances. A second issue of importance is the food security, e.g., the capacity for each state to be able to feed its population, again an issue marked by a long history of starvations. Health security has been another issue, more difficult to cope with, if only by restraining the international mobility, as experienced with the recent pandemic. Trade has rapidly become a major international relation as it allowed to improve the wealth and the well being of the populations. International finance has also been rapidly a key issue in international exchanges, be it to pay for ransoms in military wars, or to support setting up the first networks of specific trades. The progressive acknowledgment of the logic of Ricardo principle, whereby countries tend to specialize in productions where they are more efficient, taking into account the costs of transactions, helped to win the support for free trade of large parts of the populations. The long story of the above issues, be it military, trade, or migrations, made the way to cultural exchanges, soon followed in the period of industrialization, by scientific exchanges. Clearly countries were engaged at various levels and in different modes in these networks of exchanges. A first mode developed all along the phase of industrialization of the 19th century and early 20th, associating colonial trade (including in the first part of the 19thcentury trade of slaves). This “mixed” mode went on more or less till the end of the WWI. The resuming of this first wave of globalization (in Baldwin view) occurred on a radically new basis after the WWII within a world divided in three: an occidental alliance of capitalist countries (headed by the USA), an alliance of communist countries (headed by the USSR), and a third world of developing countries. A major innovation of the time was the rising importance of international organizations of the United Nations. Despite the rise of these international organizations, the international relations were still strongly marked by an inequality between developed and developing countries (cf Amin 1978), the prices of raw materials, in which developing countries were often specialized, being monitored by big firms of developed economies. This situation was somehow in contradiction with the ideal of fairness that the two alliances of the first and second world were supposed to promote, one with its ideology of Welfare State, the other with the ideology of communism. By and large this first wave of globalization supported for decades growing flows of trade. The demise of the communist alliance around 1990 and the rise of a neoliberal ideology in the occidental alliance in the 1980s led to the second wave of globalization (as stressed by Baldwin), further fueled by the diffusion of more and more sophisticated and ubiquitous technologies of information and communication technologies which allowed to relocate parts of production processes abroad. This expansion of global value chains was also strongly boosted by the rising importance of financial criteria in the strategies of firms, a shift favored by a full liberalization of financial activities. Meanwhile, this context favored the rapid growth of large developing countries, the BRICS (standing for Brazil, Russia, India, China, and South Africa), as opportunistically branded by Goldman Sachs as a highprospect security. Though, at the top of this “second” wave of globalization, at the turn of the 2010s, a major global financial crisis burst out in 2008, coming as a strong warning that the game was over, that finance should be tamed. This was concomitant with a rising conscience all 447

Pascal Petit

over the world that the “global” mode of development, experienced since the 1990s, was unsustainable. This process of an internationally rising consciousness of the environmental challenge really took off at a United Nations Conference on Environment and Development in Rio in 1992. It took though 20 years or so to lead to a general commitment of 189 countries (representing 97% of greenhouse gas [GHG] emissions) at the COP21, meeting in Paris in 2015, to voluntarily limit their emissions of GHG (by means of INDCs Intended Nationally Determined Contributions). A new era was clearly opening whereby the vision of our human relations to the world had to be revised. We had to see our future in combination with the one of our natural environment, a Galileo kind of revolution as stated. The recent pandemics of 2020 just appeared as a reminder of a safety emergency to be able to control our provision lines. Basically the turn to a sustainable mode of development has thus to be achieved within the next three decades. The recent COP26 in Glasgow in 2021 led to commitments to reach GHG neutrality by 2050 with countries or regions like the EU planning to reduce by 50% their emissions by 2030. All these commitments are clearly based on the recommendations given by contradictorily debated reports on contemporary environmental studies, the IPCCs (Intergovernmental Panel on Climate Change). The overall objective is to hold the rise in temperature below 2 degrees at least, which though implies many disturbances in climate and environment, as many countries protested in 2015 in Paris that less than 1.5 degree rise was vital for them (at least 37 Small Insular Developing Economies SIDS, now constituting a group in diverse international organizations of which UNESCO and UNCTAD). Clearly sciences are paving the way for the big paradigmatic transformation underway. It is interesting in this context to underline that economics has been more than shy in this investigation of the changing context. The first warning on the limits to growth came in 1971 from the report of a group of economists (see Meadows and alii (1972)), for the Club of Rome, funded in 1968 by a set of diverse personalities (some of the OECD) to investigate the sustainability of the mode of development of industrialized countries. But when in 2019 two eminent economists, Andrew Oswald from Warwick University and Nicolas Stern from the London School of Economics, tried to check to what extent the questions of the limits to growth and of the changes in environment had impacted the papers published by the highest-ranked economic journals, it turned out that only 57 out of 77,000 papers (e.g., 0.074%) did address these issues. How can we explain such dramatic blindness? Oswald and Stern blamed it mainly on the panurgism of the economist profession, with an obsession to publish in high ranked journals, leading to some extreme conformism. Conversely one should stress that to address the variety of issues raised by the environmental challenge interdisciplinary approaches are required (as shown by the succession of IPCC reports), an inter-disciplinarity which is most unfamiliar in orthodox economics. An evolutionary perspective is in that respect more open to grasp the undergoing transformations as it tries to take into account both changes in techniques and in learning processes. Indeed tracking the detrimental changes under view presents many aspects and all sides of production and consumption activities have to be checked thoroughly along time and spaces to see to which extent and under which conditions they contribute to the degradation of our environment. Concretely, it leads to all kinds of norms and prescriptions which have to be certified for a sustainable working of our economies. This complexity, that we qualified of paradigmatic change, sets the globalization issue in an entirely new context. The scientific discourses on the relevant norms, required to ensure sustainable modes of developments, are in this new period of transition really becoming important global public goods, in the very sense of 448

Globalization and its governance in an evolutionary perspective

Kaul et al. (2012), referring to the quality of the air we breath, of the water we drink or of the food we eat. In this context the distances between production and consumption activities which were at the core of past waves of globalization can usefully be reduced. Surely this has to be done in ways that reduce inequalities among citizens when the past waves of globalization had been associated with major increases in inequality. The discontents of the previous forms of globalization (see Stiglitz 2002) will have to be turned into supporters of the new era of international exchanges whereby the sustainability and fairness of all groups worldwide would have become a global common good. In this new big transformation, we would shift from a world where supposedly everyone tries to increase his own wellbeing (global welfare of type 1) to a world collectively trying to preserve/improve its environment while considering the wellbeing of all as a prerequisite global public good. This global welfare of type 2, driven by some kind of reflexive governance for global public goods (see Brousseau et al, 2012) could be seen as one positive outcome of innovative uses of ICT and AI, as suggested by Baldwin for a third wave of globalization. Clearly such outcome would have to follow from a major political transformation, as clearly expressed by Mireille Delmas Marty (2019) in her manifesto for a pacified globalization. In a world of some eigth billions of human being, living mainly in cities, rather well informed and enjoying powerful numerical technologies, the fatality attached to the antropocene sounds paradoxical when the SDGs (sustainable development goals) have been widely accepted and when we know the norms to which we have to adjust our activities. Three propositions can help to meet the contemporary challenge of putting the world on a sustainable trajectory: 1) preserve and value the differences, 2) acknowledge the interdependencies, and 3) promote solidarities at a world level. Though this implies a relative change of paradigm in the role of the nation states to include these goals and cooperate to achieve a global governance allowing a sustainable development at a world level. The first steps in this change of paradigm have indeed been taken, as shown in the COP meetings since 2015. But countries have taken their commitments in more or less authoritarian contexts and the above altruist propositions risk to have been neglected. As we need to speed up the process to meet the challenge of increasing environmental disorders, these differences in the depth and momentum of the paradigmatic changes can slowdown this much needed progress toward a new global governance. Even worse, these differences may fuel tensions and wars that will worsen the global environmental situation, as shown with the recent war in Ukraine, preventing the construction of an adequate global governance. A widespread concern of populations to preserve a liveable environment as well as the potentials of our information and communication technologies to keep under control the environmental infringements do constitute strong factors to help to set up a global governance respecting the three principles stated above; though the completion of this process may take too long or be met by conflicts between nation or clubs of countries. The lessons that can be drawn worldwide from the rising number of environmental disasters are crucial to accelerate the paradigmatic changes in the role of nation states needed to achieve a global governance complying with the three propositions stated above.

References Amin, Samir (1978) L’impérialisme et le développement inégal. Editions de Minuit, Paris. Baldwin, Richard (2006) Globalization: the great unbundling(s). repository.graduateinstitute.ch.

449

Pascal Petit Baldwin, Richard (2016) The great convergence: Information Technology and the New Globalization, Cambridge, MA, The Belknap Press of Harvard University. Delmas, Mireille (2019) Manifeste pour une mondialité apaisée. https://www.mediapart.fr/journal/ culture-idees/130222/le-manifeste Eric, Brousseau, Dedeurwaerdere, Tom, and Siebenhüner, Bernd (eds.), 2012 Reflexive Governance for Global Public Goods. Cambridge, MS, The MIT Press. Kaul, Inge, Grunberg, Isabelle and Stern, Marc A. (eds.) (1999). Global Public Goods: International Cooperation in the 21st Century. NY, Oxford University Press. Krassner, Stephen (2009). Power, the State, and Sovereignty. Meadows, Dennis and Donella et alii (1972) The Limits to Growth. Potomac Associates & Universe Books. Stiglitz, Joseph (2002) Globalization and its discontents. WW Norton & Company.

450

INDEX

Abbott, A. 131n6 ABM see agent-based modelling accumulation 20, 48, 129, 306, 313n16, 411 ACE see agent-based computational economics additivity 107 ad hoc problem solving 198, 200 affective psychology 93, 94–95 agent-based computational economics (ACE) 147, 148, 152; minimum intelligence by principles 152–153; minimum intelligence in practice 153–155; MinP in 147–148 agent-based economic simulations 417 agent-based modelling (ABM) 147, 359, 362, 412, 419n9 agent-based simulation 416–417 Aghion, P. 313n18, 313n21, 326 Agrawal, A. 343, 352 Ahn, J. 350 AI see artificial intelligence Akcigit, U. 326 Akerlof, G.A. 192 Albin, P.S. 150 Alchian, A. 45–46 Aldrich, H.E. 138, 313n15 Allen, R.G.D. 42 Almudi, I. 41, 47, 49, 50, 55, 60, 282n6, 390, 393 alternative facts 183n3 American institutionalism 112 anagenesis 178 analogies and metaphors see metaphors and analogies (MA) Andersen, E.S. 112 Ando, A. 344 Antonelli, C. 249, 256, 258 Appelt, S. 349

Apple 38, 201, 204, 428 aprioristic subjectivism 73–74 Araujo, R.A. 313n11 Archer, M. 232 Arndt, F. 200 Arrieta, O.A.D. 350 Arrow, K. 42, 45, 47, 49, 305, 390 artefacts 99, 229–230, 232 Arthur, W.B. 264 artificial intelligence (AI) 5, 369, 446; deep learning and 374; defined 370; as a game changer for innovation policy 371; as more than a general purpose technology 374; three essential elements of 370–37; legitimacy of stakeholders and ethical guidelines for 372–373; three essential elements of 370–371; artificial intelligence as a game change for innovation policy 371; artificial intelligence as more than a general purpose technology 374; deep learning and AI 374; visible hand as guardian of human involvement 376–379; visible hand changing direction in AI development 375–376; visible hand ensuring the legitimacy of stakeholders 375 Asch, S. 94 asset orchestration 201, 210 Atkins, W. 34 Austrian economics, F.A. Hayek 69; Carl Menger and his heirs 69; Carl Menger’s evolutionary social theory 69–71; Darwinian theory of cultural evolution of Hayek 71–73; evolutionary, naturalistic subjectivism of Hayek 74–76; Ludwig von

451

Index Mises praxiological, aprioristic subjectivism 73–74 Aversi, R. 266 Azoulay, P. 344 Babe, R.E. 81 Babutsidze, Z. 261, 266, 267, 269 Bacdayan, P. 219 Baconian programme 20 Baldwin, R. 445, 446, 447, 449 Ball, J.A. 241 Balland, P.–A. 333, 334 Banzhaf, W. 171n3 Barabási, A.L. 349 Bass, F. 266 Basurto, X. 170, 382 Bathelt, H. 351 Battiston, S. 362 Bauder, H. 344 Baum, C. 60 Bauman, Z. 392 Baumol, W.J. 187, 201, 205 Beaver, D. 345 Beck, N. 69, 76 Becker, M.C. 55, 215, 218 behavioral psychology 423 Beinhocker, E.D. 170 Bezos, J. 199, 203, 209 Bianchi, M. 270n2 Bichler, S. 38 bifurcation points 177, 183n1 “big data” revolution 200 Binz, C. 343 biological evolution 13, 26, 72, 99, 108, 109, 113, 138, 139, 144, 238, 249, 265, 423, 427 Birchenhall, C.T. 361 Blind, G.D. 167, 170, 171, 381 Blinder, A. 43 Bloch, H. 275, 276, 282n3, 282n9 Bloch, M. 113 Boltzmann, L. 151, 419n10 Boolean networks 155 Boschma, R. 332 Botta, A. 364 Boudry, M. 235, 238, 243n1 Bouis, H. 265 Boulding, K.E. 79, 80, 81, 82, 83, 84, 85, 85n2, 86n8, 154; evolutionary economics 80–82; evolutionary economics, elements of 82; government intervention 84–85; market, theory of 84; production, theory of 82–83; relevance as a founding father of evolutionary economics 79–80 bounded rationality 46, 54, 55, 108, 167, 188, 203, 227, 263, 276, 363; modelling of 262–264

Bourdieu, P. 39 Bourrat, P. 238 Bowles, S. 109 Boyd, R. 109, 138, 236, 238 Boyer, P. 404 Bozeman, B. 345 Brandeis, L. 37 Braudel, F. 113 Braun, T. 347 Bresnahan, T.F. 205 Buenstorf, G. 313n15 Burenstam-Linder, S. 191 Bush, P.D. 415 Bush, V. 343 business competence capital, nature of 187–190 capabilities 200 Cacciatorifi, E. 230 Caiani, A. 419n12 Callebaut, W. 419n8 Callegari, B. 282n10 Camic, C. 31 Campbell, D. 72 canonical economics 102, 108, 113 Cantner, U. 60, 350, 369 Capability 58, 60, 200, 434 capital 15, 35, 36 capitalism 11–12, 16, 17, 18, 24, 26, 35, 118; and socialism 22; “spirit” of 19–20 capitalist engine 17–18 Caplan, B. 406 Cardaci, A. 364 Carpenter, M.P. 345 Cass, D. 53 CAS see complex adaptive system Castiglione, F. 167 Castro, J. 167 Catalini, C. 345 cellular automaton tradition 153, 154, 155 central heuristic 107 CGE models see computable general equilibrium models Chai, A. 60, 114, 261 Chain-reaction processes 154 checkerboard model 150 Chen, P. 421 Chen, S.-H. 147 Chielens, K. 239 Chirot, D. 393 chlorofluorocarbons (CFCs) 407–408 Christiansen, M.H. 109, 236, 238 CH see continuity hypothesis Cialdini, R. 95 Ciarli, T. 60, 268, 417 Cimini, G. 350

452

Index Cimoli, M. 313n11 Cincotti, S. 419n12 Cioffi-Revilla, C. 1 citizen payoff 394 Clark, J.M. 38 climate policies, macro-evolutionary modelling of 359, 361–362; bounded rationality, climate policy under 363; distributional issues 363–364; input-output structure and resource scarcity 364–365; research gaps and issues for future modelling 363 cliometrics 184n8, 184n9 Coase, R H. 185, 186, 190, 192–193 Coccia, M. 348 Coenen, L. 130 co-evolution and economic development 285–286 co-evolutionary process 4 coevolution of innovation and demand 284; co-evolution and economic development 285–286; evolution of demand since the industrial revolution 287–288; innovation and economic development 286–287; new sectors, coevolution and economic development by the creation of 288–294 cognitive psychology 90–91 cognitive rules 384 Cohen, M.D. 219 Collingridge, D. 373 commercialization and scale up 191 commodity space 382 Community Innovation Surveys (CIS) 257 comparative evolutionary analysis 174–176 competence 187 competence bloc theory 190–193 competent customers 191 competent team, firm as 193–194 competition and pricing mechanism 423–424 competitiveness 299, 300, 301 competitiveness policy 5, 299, 300, 305 complex adaptive system (CAS) 424 complexity theory (CT) 137, 140–141 computable general equilibrium (CGE) models 364 computational evolutionary economics 147; backgrounds and motivation 147–149; hierarchies 148–149; maximum entropy principle (MEP) 151; minimum description length 149–150; minimum intelligence (MI) in agent-based computational economics (ACE) 152–155; minimum principle 147–148; neural nets and minimality 155–156; simplicity principle (SP) 149–151 computational simulation 417 connectionism 155 conservative sense, Marx interpreted in 17

Consoli, D. 263, 284 consumer demand and structural change 268 consumption: accounting for consumption laws 265; ECS behavioural theory of 102–104 contingency 178 contingency analytic approach 174 contingency graph 176–178, 183n7 contingency in evolutionary economics 174; causality relationships between different states of a process 178–179; comparative evolutionary analysis 174–176; contingency analysis approach 182; contingency approach and causal-logical terms “necessary” and “sufficient” 181; extension of contingency analysis by probabilities 181–182; graphical-analytical contingency concept 175–178; path dependency in contingency graph 179 continuity hypothesis (CH) 113, 137, 139–140 conversational evaluation 81 Cordes, C. 109, 140, 265, 313n15 Corley, E. 345 Cosmides, L. 90 country embeddedness, determinants of 349–350 COVID-19 pandemic 59, 126, 345, 352 Cowan, R. 266 Cowles commission 42 Cowling, K. 313n13 creative actions, problem solving through 231–232 creative response 252; path dependence of 253–255 Crespi, F. 256, 258 cultural entrepreneurs 12, 19, 20 cultural pessimism 24 culture 19, 24, 98, 102, 104; and democracy 24–25; of innovation 20 customer competence 191 Cyert, R.M. 96, 206, 385 Czaika, M. 348 D’Adderio, L. 229, 230 Darwin, C. 11, 13, 14–16, 25, 26, 27n12, 147 Darwin and Darwinian ideas in economics, Schumpeter on: evolution 15; Schumpeter’s “Monroe doctrine” 14 Darwinian Evolutionism 14 Darwinian principles 44, 136, 138 Darwinism 15–16, 111, 138, 139 Darwin’s ideas 136; complexity theory as an improvement of 140–141; continuity hypothesis (CH) 139–140; generalized Darwinism 138–139; heuristics and epistemological principles implied in each position 141–144; metaphors and analogies 137–138

453

Index Darwin’s theory of species evolution 426 Darwin’s variational theory 414 Darwinian theory of cultural evolution 71–73 David, P.A. 251, 320 Davies, B. 334 Dawid, H. 167, 419n12 Debreu, G. 42 decision makers 188 decision-making process 227 de-coordination 164 dedicated innovation systems 60 deductive formats 381, 385–386 Deegan, J. 335 deep learning 370; and artificial intelligence 374; visible hand as guardian of human involvement in the era of 376–379 Delli Gatti, D. 419n12 Delmas, M. 449 demand and innovation 267–268 demand and supply, coevolution of 268–269 democracy 25 democracy and its paradoxes 391; expertise vs direct participation paradox 392; novelty vs retained practice paradox 392–393; representativeness vs governance paradox 391–392 democracy as an evolutionary process 393; citizen payoff 394; emergent properties 395–397; inter-subsystem dynamics and coevolution 395; intra-subsystemic evolution 394–395 Denis, B. 230 Dennett, D.C. 235, 238, 239 Denzau, A.T. 406 De Rassenfosse, G. 168 developmental turn 128 de Vries, H. 13 Diamond, P. 48, 53 Dierker, E. 42 digital and energy transformation 436–439 digital communications, infrastructural investments for 439–440 digital data analytics 435 digital economy 439 digitalization 434 direction of change 118, 328 discovery 16 dispositions 75–76 distributed capacities 328 Dittrich, P. 171n3 division of labor 421; competition and pricing mechanism 423–424; to co-ordination of innovation 428; culture and institutions 426–427; evolution of organization and classification of firms 426; money,

changing nature of 425–426; price, structure, and knowledge 424; science philosophy, issues in 427–428; trade–off between stability and complexity 422–423 docility 227–228 Dopfer, K. 1, 11, 12, 26n1, 60, 90, 108, 109, 110, 113, 162, 165, 167, 170, 171, 171n1, 183n7, 190, 192, 239, 251, 258n1, 269, 312n1, 381, 385, 387, 388 Dopfer-Potts Generalized Rule Approach 386 Dorfman, R. 45 Dosi, G. 51, 52, 55, 56, 59, 60, 61, 167, 231, 361, 362, 417 Dosi-type of models 360 Duffy, J. 150, 153 dynamic capabilities 56, 304; contributions to 206–207; and evolutionary economics reimagined 209–210; innovation and change, narrow view of 207–208 dynamic performance of capitalism and socialism 22 Earl, P.E. 96, 264 economic development 6, 18, 102, 108, 111, 126; co-evolution and 285–286; innovation and 286–287; micro, meso and macro levels of 302–304 economic history, evolutionary economics and 107; history of history 110–111; mutual considerations 111–114; ontological and heuristic foundations of 107–110 economics of smart specialisation policy 330 EC see European Commission ECS see evolutionary cultural science Edler, J. 344 Edquist, C. 141 EEG see evolutionary economic geography EEP see evolutionary economic policy Eiconics 154 Eliasson, G. 185, 189, 194n2 El Qaoumi, K. 264 embeddedness and performance 350–351 emergent properties 141, 395–397 EME see evolutionary macroeconomics Energy Sector Strategic Plan (2018) 440 ENGAGE model 361 Engel, E. 265 Engel Curves 268 Engel’s law 265 Enos, J. 321 entrepreneurial competition, nature of 186–187 entrepreneurial discovery process 320, 328, 325, 334; aggregation level 326; spillovers 326; strategic complementarities 325–326 entrepreneurial management 202–204, 205, 206

454

Index entrepreneurship 201–202, 203, 205; multistakeholder support for entrepreneurship ecosystem 440–441; Schumpeter’s concept of 424 entropy 151 Environmental Protection Agency (EPA) 407 EOE see Experimentally Organized Economy EPA see Environmental Protection Agency EPE see evolutionary political economy EPO see European Patent Office equifinality 178 Erhard Miracle 425 ethical guidelines of expert councils 373 EURACE-model 361 Eurocentric concepts 110 European Commission (EC) 317 European Patent Office (EPO) 168 evolutionary computation paradigm 147 evolutionary consumer theory 261; consumer demand and structural change 268; consumption laws, accounting for 265; demand and innovation 267–268; demand and supply, coevolution of 268–269; modelling bounded rationality 262–264; population thinking 265–267 evolutionary cultural science (ECS) 98; behavioural theory of consumption 102–104; culture of economics 104; evolutionary economic geography (EEG) 117, 332; contribution of 119–125; evolution of 123–130; future challenges and directions 123–130; geographical foundations of economy 117–119; key guiding principles 119–122; key spatial policy issues 129–130; moving beyond ‘patchwork’ evolutionary economic geography 127–129; putting the principles in place 122–125; scope of 120, 119–125; and smart specialization policy 333–334; smart specialization policy, design and implementation of 334–337; widening the field 125–127 evolutionary economic policy 1, 4–5, 6 evolutionary economic policy and competitiveness 299; evolve, ability to 306–308; failure, rationalities of 305–306; integrated classification 308–310; Krugman’s critique 301–302; micro, meso and macro levels of economic development 302–304; qualitative change and growth ‘beyond GDP’ 300–301; system functions 305–308 evolutionary epistemology 31, 72 Evolutionary Game Theory 44 evolutionary macroeconomics 1, 3, 4

evolutionary market theory 4 evolutionary political economy 1, 5, 6 evolutionary price theory 275; prices in motion 277–279; prices in orderly markets 276–277; prices in the economy as a whole 279 evolutionary psychology 92–93 evolutionary-structuralist agenda 303 evolutionary-structuralist perspective 300 evolutionary subjectivism 74–76 evolutionary thinking 16, 79, 206 Experimentally Organized Economy (EOE) 186–188 experimentally organized firm, rational foundation of 185–186 expertise vs direct participation paradox 392 facts, ideas vs. 18–19 Fagerberg, J. 433 falsifiability and ongoing evolution 427–428 Fatas-Villafranca, F. 41, 60, 390 Feldman, M.S. 229, 232 Fellner, W. 270n2 Fester, R. 148 firm as an experimental decision maker 185; business competence capital, nature of 187–190; commercialization and scale up 191; competence bloc 193; competence bloc theory 190–191; competent customers 191; as competent team 193–194; entrepreneurial competition, nature of 186–187; experimentally organized firm, rational foundation of 185–186; extreme heterogeneity 189–190; receiver competence 191 firm heterogeneity 125 fiscal policy 310, 312, 425–426 Fisher, R.A. 44 Fisher’s Principle of evolutionary change 275 Fogel, R. 178 Fogg, P. 175 Foray, D. 316, 338 force field analysis 93 Foster, J. 90, 140, 141, 142, 143, 144n1, 167, 239 foundational evolutionary traverse 51–55 fourth age of research 342 Frame, J.D. 345 Frank, R. 39 Freeman, L.C. 58, 60, 347, 434 Freeman, R.B. 346 Frenken, K. 266 Friedman, M. 43, 46, 53 Frisch, M. 175 Frisch, R. 42 Fudenberg, D. 58

455

Index Galbraith, J.K. 38 Gale, D. 45 Gallegati, M. 59 Galunic, C. 240 Game Theory 44, 52 Gantner, A. 257 Gatti, D.D. 167 Gazni, A. 348 Gehrke, C. 27n8 Geisteswissenschaft 99, 100 general equilibrium theory (GET) 42, 45 Generalized Darwinism (GD) 119, 127, 136, 138–139, 162 Generalized Rule Approach 382–383, 385, 386 General Theory of Economic Evolution 162 genetic programming 148–149 Georgescu-Roegen, N. 270n2, 412, 416 Gerdes, L. 419n11 GET see general equilibrium theory Gigerenzer, G. 153 Ginsburg, S. 236 Gintis, H. 109 globalization 445–449 global knowledge embeddedness 341; collaboration and mobility, knowledge diffusion through 342–343; country embeddedness, determinants of 349–350; embeddedness and performance 350–351; from individual interactions to a global structure 346–347; international collaboration, determinants of 345–346; international mobility and collaboration, drivers of 343–346; scientist mobility, determinants of 343–345 global knowledge network 347–349 Glänzel, W. 346 Gode, D.K. 152, 153 Goldberger, A. 43 Gould, S.J. 183n2, 415 government intervention 84–85 Graf, H. 341, 350 Grapard, U. 162 graphical-analytical contingency concept 175–178 great divergence 20, 104 Great Recession 59 Grebel, T. 161, 165, 174 Greenwald, D. 38 Griskevicius, V. 270n3 Gruber, J. 330n3 Gui, Q. 348, 349 Gul, F. 108, 113 Haas, D. 278, 282n9 Haavelmo, T. 42

habit psychology 91–92 habituation 92, 264 Haddad, L. 265 Hafner, S. 361 Hahn, F. 42 Hanappi, H. 418n1 Harvey, D. 418n2 Hausmann, R. 321 Hayek, F.A. 42, 43, 308, 392, 424; evolutionary, naturalistic subjectivism 74–76 Heath, D. 95 Hebb, D.O. 155 Hecker, A. 257 Hegel, G.W.F. 15 Heilbroner, R. 81, 82, 85 Helfat, C.E. 198 Henning, M. 119 Henrich, J. 109 Herrmann-Pillath, C. 38, 92, 98 heterogeneity 108, 118, 121, 279 heuristics 141–144 Heylighen, F. 239 Hicks, J.R. 42, 43 hierarchy hypothesis 148 Hirschman, A.O. 322 Hirshleifer, J. 326 Hodgson, G.M. 30–31, 55, 90–92, 117, 127, 139, 183n5, 217, 220, 240, 243n5, 313n15 Hofbauer, J. 44 Hofhuis, S. 235 Hofhuis, S.T. 235 holism 121, 122 homogeneity 107, 150 homo oeconomicus 18, 107, 111, 161 homophily principle 156 homo sapiens oeconomicus 162, 165, 170 Hornung, E. 344 Hou, L. 349 Howitt, P. 313n18, 313n21 Hudson, M. 35 Hull, D.L. 138, 217 human history 110 Huxley, J. 418n7 Ida, T. 269 ideas vs. facts 18–19 income distribution and social justice 22–23 income distribution policy 24 industrially competent venture capitalists 191 Industrial Revolution 12, 20, 103, 287–288 “Industry 4.0” revolution 126 informationalists 238 information and communication technologies (ICTs) 319–320, 345, 392, 435, 436, 439–440, 445–446, 449

456

Index infrastructural investments 439–440 innovation 435; “culture” of 20; and economic development 286–287; path dependence and persistence of 255–258 innovation policy 369; legitimacy of stakeholders and ethical guidelines for 372–373 innovation success, degree of 54, 256 innovation systems 112, 335 institutions as social rules 163–164; culture and institutions 426–427 Integrated Assessment Models (IAMs) 359, 362–363, 364 Intergovernmental Panel on Climate Change (IPCCs) 448 international collaboration 342–343, 348–351; determinants of 345–346 international patent classification (IPC) 168 inter-subsystem dynamics and co-evolution 394–395; formal analysis for 400–401 IPC see international patent classification IPCCs see Intergovernmental Panel on Climate Change Isaksen, A. 129 Jablonka, E. 236 Jackson, M.O. 303 Jackson, T. 270n3 Jaffe, A. 319 James, W. 92 Janssen, M. 417 Jaynes, E.T. 151 Jeong, S. 346 Jevons, W.S. 41, 161 Jobs, S. 201, 204 Johnson, S. 330n3 Jones, B. 319 Jones, C.I. 312n6 Jöns, H. 344, 346 Ju Jung, E. 231 Kahneman, D. 55, 89, 201 Kalthaus, M. 341, 350 Kato, M. 344 Katz, J.S. 345 Kauffman, S.A. 155, 241 Kaul, I. 449 Kelly, G.A. 93 Kermack-McKendric Model 154 Kesting, S. 79 Keynes, J.M. 22, 23, 43 Keynesian economics 422, 426 Keynesian-Neoclassical synthesis 43 Keynesian policies 284 Khalil, E. 80–81 Kiopa, A. 345

Kirman, A. 59 Kirzner, I.M. 404 KISS principle 149–150, 153 Klein, B. 49 Klein, L. 43 Klenow, P.J. 312n6 knowledge, path dependence in the generation of 252–253 knowledge space 382 Knudsen, T. 55, 217, 218, 220 Koestler, A. 240 Kolesnikov, S. 349 Kolmogorov complexity 149 Koopmans, T. 42, 45 Kozo-Polyansky, B.M. 148 Kraemer-Mbula, E. 433 Krueger, A.O. 313n12 Krugman, P. 118, 300, 301–302 Kulturwissenschaft 99 Kurz, H.D. 11, 26n2, 27n6, 27n8, 282n9 Kuznets, S. 42 Lades, L.K. 265 Lamoreaux, N. 36 Lamperti, F. 362, 419n13 Larivière, V. 342, 349 Lavoie, M. 270n3 “laws of motion” of capital accumulation 128 “Law” of the falling rate of profit 17 Lazaric, N. 226, 230 LDCs, evolutionary economics and 433; digital and energy transformation in Kenya and Rwanda 436–439; digital communications and renewable energies 439–440; digital economy, governance and regulatory environment for 439; entrepreneurship ecosystem, multi-stakeholder support for 440–441; low-income countries, economic development in 433–435; skills development, policies for 440; transformative change, policies for 439–441; transformative change in Kenya and Rwanda 435–439 leadership 49, 58, 205 Le Doux, J. 95 Leeuwen, T.N. van. 346 Lehmann-Waffenschmidt, M. 183n4 Leih, S. 203 Leung, R.C. 346 Levins, R. 414 Levinthal, D.A. 209, 219, 229 Levi-Strauss, C. 232 Levit, G.S. 127 Lewin, K. 93 Lewontin, R. 414

457

Index Leydesdorff, L. 348, 349 Li, P. 351 Lima, G.T. 313n11 Lindberg, T. 189 Ljungqvist, L. 58 lock-in situations 371, 374 longue durée 104, 113 Lorentz, A. 268 Lorenz, E. 433 low-income countries, economic development in 433–435 Lucas, R.E. 49 Lyell, C. 147 MA see metaphors and analogies machine learning (ML) 371 Macroeconomic Keynesian aggregate models 45 macro-economic models 360 macroeconomic welfare-optimal equilibrium 107 macro-evolutionary modelling of climate policies 359, 361–362; bounded rationality, climate policy under 363; distributional issues 363–364; input-output structure and resource scarcity 364–365; research gaps and issues for future modelling 363 macro-evolutionary models 360, 361, 363, 364 Maintain or Improve Profit (MIP) targets 187 Malerba, F. 50, 51 Malthus, T.R. 20 Manig, C. 265 maps 416–417 March, J. 91, 96, 228 March, J.G. 206, 385 Margulis, L. 148 market, theory of 84 market-share competition 422 Markey-Towler, B. 89, 278, 282n6 Marmefelt, T. 86n10 Marrocu, E. 335 Marschak, J. 46 Marshall, A. 16, 27n4, 117, 131n2, 161, 288 Martin, B.R. 345 Martin, R.L. 117, 128 Marx, K. 6, 11, 15–16, 17–18, 19, 20–22, 27n7, 41, 413, 423; capitalist engine 17–18; in a “conservative sense” 17; “Law” of the falling rate of profit 17 Marx-Schumpeter-Veblen complex 415 Maré, D.C. 334 Maslow, A.H. 150 Massey, D. 122 materiality 101 Mathews, J.A. 231 maximum entropy principle (MEP) 151 Maynard-Smith, J. 44

McCloskey, D. 85n2 McCraw, T. 26n2 McCulloch, W.S. 156 McCulloch-Pitts neural net (MPNN) 155, 156 McFadden, D. 45 Meadows, D. 448 Melkers, J. 345 memes 235; creativity and innovation from perspective of 241–242; cultural evolution, imitation, and eye view of 236–238; as information and instruction 238–239; interconnection 239–240 memetics 101 Mendel, J.G. 13 Menger, C. 41, 270n3, 288; evolutionary social theory 69–71; and his heirs 69 mergers and acquisitions (M&As) 186 Meso-centred Schumpeter-Veblen dynamics 417 Mesoudi, A. 109 metaphors and analogies (MA) 136–139 Metcalfe, J.S. 11, 55, 58, 137, 275, 278, 279 Metcalfe, S.J. 275, 276, 313n15 Methodenstreit 111 methodological dualism 74 methodological individualism 170 Miao, J. 58 micro, meso and macro levels of economic development 302–304; “micro-mesomacro” synthesis 90 micro-foundations 399–400; micro-unit 2 Microsoft 38 Mill, J.S. 41 minimality, neural nets and 155–156 minimum intelligence (MI) in agent-based computational economics (ACE) 147, 152; in practice 153–155; by principles 152–153; minimum principle and description length (MDL) 147–151 Minsky, H.P. 418n4 Mises, L. 43, 73–74, 393 Mittelstadt, B. 373 models 416–417 Mokyr, J. 19, 109 Moneta, A. 265 monetary policy 309–310 money, changing nature of 425–426 Montgomery, S. 393 Moser, P. 344 MOSES model 189, 194n3 motion, prices in 277–279 Mott, T. 80, 86n8 Mueller, M. 269 multidimensional competition 185 multi-stakeholder support for entrepreneurship ecosystem 440–441

458

Index Musk, E. 202 mutations 13 Muth, J.F. 52 Muñoz, F.F. 60 Myrdal, G. 117 Müller, G. 418n7 Narin, F. 346 narrative economics 154 Nash, J. 44 natural evolution 139 naturalistic, evolutionary subjectivism 74–76 naturalistic and bioeconomic perspectives 90, 92–93 natural selection 15–16 Nelson, R.R. 1, 11, 42, 46, 48, 50, 61, 90–91, 108–109, 131n1, 136, 138, 167, 197–199, 201, 206–209, 215–216, 218, 220–222, 228–229, 232, 263, 275–278, 284, 299, 313n15, 313n21, 360 Nelson and Winter, foundational evolutionary traverse of 41; economic order and firm theory in the early Sidney Winter 45–48; far beyond the consolidated beachhead 59–61; foundational evolutionary traverse 51–55; golden decades 56–59; problem of economic change in the early Richard Nelson 48–51; Nelson–Winter collaboration 50Neodarwinian synthesis 101 neoinstitutional economics 112 neo-Schumpeterian approaches 108–109; and tradition 90–91 neo-Walrasian general equilibrium theory 42 network configurations 37 neural nets and minimality 155–156 new economic geography (NEG) 118 Newell, A. 90, 91, 227 new sectors, coevolution and economic development by the creation of 288–294 Nguyen, T.V. 350 niche markets 267 Nisticò, S. 270n2 Nitzan, J. 38 NK models 208, 209 North, D.C. 112, 406, 407 novelty vs retained practice paradox 392–393 Nunn, N. 109 Nuvolari, A. 51 Occam’s razor 148 OECD 369 O’Hara, P.A. 415 Ontology and evolutionary methodology 98, 100, 102

Oosterlynk, S. 129 Open loop 174 OPE see Original Political Economy Orazbayev, S. 348 orderly markets, prices in 276–277 organic understanding 70 organizational capability 56 Organizational Ecology 162 organizational routines 226, 228–229; artefacts as mediators between skills and routines 229–230; cognition, problem solving, and routines 227; docility 227–228; Nelson and Winter’s observation of 228–229; new patterns to build and new challenges to handle 231; problem solving through creative actions 231–232; replication and emergence of new routines 229 Original Political Economy 6 Orlikowski, W.J. 230 orthogenesis 178 Ortmann, G. 183n2 Ostrom, E. 5, 111, 170, 239, 381–383, 385, 386, 388, 402–405, 407, 417 oversaturation 294n3 Panksepp, J. 95 panmemetics 235 “parasitical” memes 237 Pareto, V. 161 Parmentier-Cajaiba, A. 230, 231 Pasinetti, L.L. 43, 304, 313n11 Patalano, R. 80 path dependence 249; of creative response 253–255; in the generation of knowledge 252–253; and persistence of innovation 255–258 Patrucco, P.P. 249 Paul, J. 175 Paulin, J. 167 Pay-As-You-Go business model 437 Peneder, M. 299, 312n1, 418n3 Penrose, E.T. 304 Pentland, B.T. 229, 231, 232 Perez, C. 434 perfect rationality 262 personality psychology 93–96 Pesendorfer, W. 108, 113 Peteraf, M. 203 Peters, B. 257 Petersen, M.B. 404 Petit, P. 445 Peukert, H. 30 Phelps, E.S. 47, 326 Pierce, L. 200 Pigliucci, M. 418n7

459

Index Pinheiro, F.L. 336 Pinker, S. 90 Pisano, G. 197 Pitts, W. 156 Plotnikova, T. 349 Pluralism 131n8 Podolny, J.M. 208 Pohl, H. 349 Polanyi, M. 43 Poledna 365 political economy 412–416 Ponzi games 57 Popper, K.R. 43, 72, 111, 114, 392 population thinking 265–267 Porcile, G. 313n11 Potts, J. 1, 11, 60, 90, 113, 162, 167, 170, 171, 171n1, 239, 242, 382, 383, 384, 385, 387, 388 Powell, W.W. 341 praxeological, aprioristic subjectivism 73–74 Price, G.R. 44 prices: in the economy as a whole 279–280; in motion 277–279; in orderly markets 276–277 private equity market 191 private property 33 probability-enhanced contingency analysis 181 problem solving through creative actions 231–232 production, theory psychology, evolutionary economics and 89; affective psychology 94–95; existing foundations 90; growth, opportunities for 93; naturalistic and bioeconomic perspectives 92–93; neo–Schumpeterian perspectives 90–91; personality psychology 93–94; persuasion psychology 95–96; social psychology 94; Veblenian perspectives 91–92 Public Choice theory 405 public entrepreneurship 402–404; and economic evolution 406–408; political economy of 404–406 Pyka, A. 1, 60, 141, 167, 170, 269, 292, 293, 313n21 QWERTY economics 251, 259n2 R&D activities 360, 362 R&I policy 376 Radner, R. 45, 46 Rainer, A. 414 Rake, B. 349, 350 random effects (RE) probit model 257 Rasskin-Gutman, D. 419n8 RBA see rule-based approach receiver competence 191

re-coordination 164 regional differentiation 328 Reinert, E.S. 131n3 reinforcement learning 154 relatedness 124 related variety 124 relational density 328 renewable energies, infrastructural investments for 439–440 Rengs, B. 364, 415, 417, 419n11, 419n13 representativeness vs governance paradox 391–392 Rerup, C. 229 Resch, A. 107, 418n3 resilience 125 resource partitioning 304 Responsible Research and Innovation (RRI) systems 370, 373, 377–378 retention 218–221 Revay, P. 167 revealed comparative advantage (RCA) 302 Reza, Y. 175 Ribeiro, L.C. 349 Ricardo, D. 41 Riccaboni, M. 348, 351 Riccetti, L. 419n12 Richerson, P.J. 109, 138, 236, 238 Rigby, D. 330n1, 334 Rissanen, J. 151 Robert, V. 109, 136 Rodrik, D. 321 Rosenblatt, F. 156 Rothblum, U. 55 routines 198–201, 215; as replicators and firms as interactors 216–217; retention 218–221; role of 215–216; selection of firms and selection of 218; variation, selection, and retention, and their interaction 222; variation of 221–222 Roventini, A. 60 RRI systems see Responsible Research and Innovation systems rule-based approach (RBA) 161, 239; benefits of the taxonomy 169–170; economic evolution 164–165; empirical quest 167–169; generic rules and trajectories 165–167; optimization 161–162; Robinson Crusoe as a rule-user and-maker 162–164; instituiones as object rules 163–164 rule retention 4 Rural Electrification and Renewable Energy Corporation (REREC) 440 Ruse, M. 15 Russo, A. 364 routines 5, 383–385; organizational routines 228–229

460

Index Saad, G. 270n3 Safarzynska, K. 359, 362, 419n13 Salvato, C. 230 Sampat, B.N. 199 Samuels, W.J. 81 Samuelson, P.A. 42, 43 Saraceno, F. 364 Sargent, T. 58, 153 SARS-CoV-2 see severe acute respiratory syndrome coronavirus type 2 saturation 289, 294n2 Saviotti, P.P. 141, 263, 269, 284, 292, 293 Saxenian, A. 343 Scellato, G. 256, 344 Schlaile, M.P. 235, 243n1 Schmookler, J. 267, 284 Schnellenbach, J. 402 Scholz-Wäckerle, M. 411, 415, 417, 418n1, 419n11 Schubert, A. 347 Schumpeter, J.A. 11, 26n3, 27n5, 27n7, 27n10, 27n12, 43–44, 49, 84, 90, 111, 118, 161, 165, 201–202, 207, 253, 269, 275, 276, 277, 280, 281, 282n5, 282n10, 282n3, 284, 288–289, 299, 300–301, 304, 310, 312n4, 313n18, 372, 390, 404, 413–414, 433–434; capitalist engine 17–18; Charles Darwin and biological evolution 13; concept of entrepreneurship 424; culture 24; “culture” of innovation 20; Darwinism 15–16; democracy 25; evolution 15; ideas vs. facts 18–19; “Law” of the falling rate of profit 17; Marx interpreted in a “conservative sense” 17; Schumpeter vs. Marx 20–21; Schumpeter’s “Monroe doctrine” 14; social conflicts 21–24; “spirit” of capitalism 19–20 Schumpeterian catalytic R&I policy 375, 376 Schumpeterian competition 275; rivalry 256 “Schumpeter-meeting-Keynes” generation 59 Schumpeter on Darwin and Darwinian ideas in economics: evolution 15 Schumpeter on Engels and Marx; on Darwinism 15–16; discovery as important as that of heliocentric system 16 Schumpeter vs. Marx 20–21 Schuster, P. 44 Schütz, M. 414 science philosophy, issues in 427–428 scientific socialism 16 scientist mobility, determinants of 343–345 Scott, R. 80, 84–85 search rules 385 Seidl, R. 270n2 self-contamination 31 self-discovery 376

self-referentiality 31 self-talk 232 Selten, R. 153 Sen, A. 45 severe acute respiratory syndrome coronavirus type 2 (SARS-CoV-2) 238 Shannon, C.E. 151 Shiller, R.J. 192 Shuen, A. 197 Sigmund, K. 44 Silverberg, G. 55 Simon, H. 48, 55, 91, 148, 149, 188, 226, 227–228, 419n8 simplicity principle (SP) 149; maximum entropy principle (MEP) 151; minimal principle (MinP) 150–151; minimum description length 149–150 skills and routines, artefacts as mediators between 229–230 skills development, policies for 440 smart specialisation strategy (S3) 316; aiming at boosting both vitality and inclusion 326–328; concentration 318–319; differentiation 318; economic geography of innovation 318–320; entrepreneurial discovery process 325–326, 328–329; history of 317–318; regional innovation policies 320–321; the new geography of innovation 319–322; planning and entrepreneurial discovery process 320; smart specialization policy: design and implementation of 334–337; as part of evolutionary economic geography 333–334 Smith, A. 19, 21, 24, 26n3, 33, 41, 102, 153, 421, 422–423, 425, 428 Smith, K. 313n13 Smith-May-Chen (SMC) Theorem 423 social conflicts: confounding capitalists and entrepreneurs 21–22; dynamic performance of capitalism and socialism 22; income distribution and social justice 22–23; income distribution policy 24; stabilisation policy 23 social epidemics 154 socialism 25 social network effects 103 social psychology 93, 94 social rules 385; as object rules 163 social tools 72 socioeconomics 154 Solow, R. 43, 241 species evolution, Darwin’s theory of 426 Spencer, H. 15 “spirit” of capitalism 19–20 spurious persistence 255

461

Index Sraffa, P. 282n9, 425 stabilisation policy 23 stabilization 164 stakeholders and outsourcing of ethical issues of AI 372–373 steady state growth 307 Stiglitz, J.E. 312n6 Stilgoe, J. 373 stochastic processes, theory of 47; stochastic variation 307 Stoelhorst, J.W. 138 strategy-making processes 210 Strohmaier, R.M. 170 structural change, consumer demand and 268 structural economic dynamics 304 Stuart, T.E. 208 Sturn, R. 26n2 Sunder, S. 152, 153 Sunley, P.J. 117, 128 surveillance capitalism 38 symbolism 155 Szulanski, G. 220 Tarde, G. 241, 243n4 technological revolutions 434 technology races 302 Teece, D.J. 56, 197, 203, 207; Teecian dynamic capabilities 200 “teleological” processes 178 TEVECON model 285, 288–289, 293, 294n1 Thaler, R. 89 Theil, H. 42 theory construction 108 Theory of the Firm 45 Tinbergen, J. 42 Tirole, J. 58 Tit-for-Tat strategy 150 Tobin, J. 43, 48 Tomlinson, P.R. 313n13 Tooby, J. 90 Total Factor Productivity (TFP) 257 Toyota’s lean production model 200 Trajtenberg, M. 327 transformational evolutionary system behaviour 417 transformational road map 326 transformative change, policies for 439; governance and regulatory environment for digital economy 439; infrastructural investments for digital communications and renewable energies 439–440; in Kenya and Rwanda 435–439; multi-stakeholder support for entrepreneurship ecosystem 440–441; skills development, policies for 440

tribalism 392 Trigg, A. 313n11 Truffer, B. 343 Trump, D. 39 Tsouri, M. 351 Tullock, G. 313n12 tâtonnement process 277 uneven geographical development 118 universal Darwinism 138 unrestricted inequalities 24 Urban, T. 202–203 Valente, M. 60, 262, 264 Valentinov, V. 82, 84 Vanberg, V.J. 69 van den Bergh, J. 419n13 Varian, H.R. 45 Veblen, T. 27n4, 30, 41, 90, 91, 92, 112, 161, 221, 288, 292; applied evolutionary economics 32–34; evolutionary heterodox supply-side economics 34–37; as a precursor of evolutionary economics 37–39; theoretical approach and evolutionary epistemology 30–31 Veblenian perspectives and model 90–92 Veit, W. 235, 240, 243n1 Velikovsky, J.T. 240 venture capitalists (VCs) 193 Verginer, L. 348, 351 visible hand: changing direction in AI development 375–376; ensuring the legitimacy of stakeholders 375; as guardian of human involvement in the era of deep learning 376–379 von Bülow, C. 235, 243n3 von Mises, L. 161 Von Neumann, J. 148, 150, 155, 156 Wagner, C.S. 345, 346, 348, 349, 350, 352 Waller, W. 81 Walras, L. 41, 161, 288 Waltman, L. 345 Wang, J. 344, 348 Warglien, M. 219 Waters, R. 80, 82, 83, 86n7 “wear and use” effects 252 Weber, M. 19, 24 Weeks, J. 240 Welfare Economics 45 Werker, C. 369 Westphalian order 446 Wieser, F. 301, 312n3, 312n5, 312n7 Wilkins, J.S. 238 Williams, G.C. 238, 239

462

Index Williamson, O.E. 47, 112 Wilson, D.S. 59, 419n8 Windrum, P. 361 Winter, S.G. 11, 42, 43, 44, 45–48, 51, 52, 53, 54, 55, 56, 58, 59, 90, 91, 136, 167, 197, 198, 199, 200, 201, 202, 203, 206, 207, 209, 210, 215, 216, 219, 220, 221, 222, 228–229, 232, 263, 275, 278, 313n15, 360 Witt, U. 60, 69, 76, 90, 92, 109, 114, 263, 313n15 Wood, J.C. 30 World Bank Group 365, 441

Wright, I. 152, 154 Yariv, L. 303 Yoguel, G. 136 Youtie, J. 349 Zeppini, P. 266 zero-intelligence (ZI) traders 152 zero-sum-games 302 Zitt, M. 349

463